Accuweather new contract

[Bob Koehler]

Accuweather has just won a contract with Verizon, taking it away
from The Weather Channel.

Last I heard, Accuweather was a big user of VMS. I wonder if this
new contract will lead to increased sales of VMS to Accuweather?

[Kerry Main]

Not sure what the latest is wrt to Accuweather and OpenVMS, but
these are some past testimonial links:

OpenVMS - Attunity:
http://tinyurl.com/VMS-Accuweather-1

OpenVMS Attunity Video testimonial
http://tinyurl.com/VMS-Accuweather-2
ftp://ftp.hp.com/pub/openvms/wmv/ACCUWEATHER.WMV

Regards,

Kerry Main
Back to the Future IT Inc.
.. Learning from the past to plan the future

Kerry dot main at backtothefutureit dot com

[Jan-Erik Soderholm]

Kerry Main skrev den 2015-03-10 18:55:
>> -----Original Message-----
>> From: Info-vax [mailto:info-vax...@info-vax.com] On Behalf Of
>> Bob Koehler
>> Sent: 10-Mar-15 1:22 PM
>> To: info...@info-vax.com
>> Subject: [New Info-vax] Accuweather new contract
>>
>>
>> Accuweather has just won a contract with Verizon, taking it away
>> from The Weather Channel.
>>
>> Last I heard, Accuweather was a big user of VMS. I wonder if this
>> new contract will lead to increased sales of VMS to Accuweather?
>>
>
> Not sure what the latest is wrt to Accuweather and OpenVMS, but
> these are some past testimonial links:
>
> OpenVMS - Attunity:
> http://tinyurl.com/VMS-Accuweather-1
>

“Our long-term plan is to keep our backend AccuWeather
Forecast Engine on OpenVMS because it has a viable future,”
comments Fiore. “HP is breathing new life into OpenVMS by
porting it to their industry-standard Integrity servers..."


The Times They Are a-Changin'...

[JF Mezei]

On 15-03-10 16:56, Jan-Erik Soderholm wrote:

> “Our long-term plan is to keep our backend AccuWeather
> Forecast Engine on OpenVMS because it has a viable future,”
> comments Fiore. “HP is breathing new life into OpenVMS by
> porting it to their industry-standard Integrity servers..."

You could update this very simply by changing "HP" to "VSI" and removing
"their" and "Integrity".

“Our long-term plan is to keep our backend AccuWeather
Forecast Engine on OpenVMS because it has a viable future,”

comments Fiore. “VSI is breathing new life into OpenVMS by
porting it to industry-standard servers..."

[VAXman-]

Jess Goodman would be a good candidate to comment on this. I'm pretty sure
he lurks and reads c.o.v.

--
VAXman- A Bored Certified VMS Kernel Mode Hacker VAXman(at)TMESIS(dot)ORG

I speak to machines with the voice of humanity.

[seasoned_geek]

On Tuesday, March 10, 2015 at 12:22:43 PM UTC-5, Bob Koehler wrote:
> Last I heard, Accuweather was a big user of VMS. I wonder if this
> new contract will lead to increased sales of VMS to Accuweather?

It has been a few years since I received a phone call about working there. Those calls are typically pitching some extremely low paying work. They are "near" a college and the primary draw or pitch for them was people who had graduated from that school and wanted to return to the area/environment.

Very nice people to speak with on the phone, though.

It is something to keep in the back of one's mind for the social security years.

[Bill Gunshannon]

In article <mdnlp3$lpt$1...@news.albasani.net>,

Can anyone even verify that Accuweather is still using VMS?
That testimonial, like the others, is ancient history.

bill


--
Bill Gunshannon | de-moc-ra-cy (di mok' ra see) n. Three wolves
bill...@cs.scranton.edu | and a sheep voting on what's for dinner.
University of Scranton |
Scranton, Pennsylvania | #include <std.disclaimer.h>

[clairg...@gmail.com]

Yes, Accuweather still uses VMS. We started talking with them at last fall's Boot Camp.

[VAXman-]

Well, I can state that Jess Goodman was in attendance at the last OpenVMS
bootcamp. I can't imagine Accuweather sending him there if they were not
still running on VMS.

[David Froble]

And here I was under the impression that the SS years was for building
airplanes ....

What am I to do with the projects in the shop ????

[Bill Gunshannon]

In article <00AF3F7A...@sendspamhere.org>,

Might want to talk with him again. In the past week I have seen numerous
commercials lately claiming Accuweather is based on "The Microsoft Cloud".
I doubt that is VMS based.

[VAXman-]

Well, it wasn't M$ bootcamp he was attending. ;)

[Bob Koehler]

In article <cnahfk...@mid.individual.net>, bi...@server3.cs.scranton.edu (Bill Gunshannon) writes:
>
> Might want to talk with him again. In the past week I have seen numerous
> commercials lately claiming Accuweather is based on "The Microsoft Cloud".
> I doubt that is VMS based.

I suspect they have PCs on desks. And I suspect that is not thier
compute engine for weather forcasting.

But VMS can play in an MS authentication and access MS servers, so
I suspwect it could be convinced to access either of those "in the
Cloud" since the one established fact of Cloud technology is its
nebulousness.

[Bill Gunshannon]

In article <XApkMI...@eisner.encompasserve.org>,

koe...@eisner.nospam.decuserve.org (Bob Koehler) writes:
> In article <cnahfk...@mid.individual.net>, bi...@server3.cs.scranton.edu (Bill Gunshannon) writes:
>>
>> Might want to talk with him again. In the past week I have seen numerous
>> commercials lately claiming Accuweather is based on "The Microsoft Cloud".
>> I doubt that is VMS based.
>
> I suspect they have PCs on desks. And I suspect that is not thier
> compute engine for weather forcasting.

"The Microsoft Cloud" is not "PCs on desks". This whole "cloud" thing
is being pushed as the ability to have massive compute power without
having to buy the compute engines themselves.

>
> But VMS can play in an MS authentication and access MS servers, so
> I suspwect it could be convinced to access either of those "in the
> Cloud" since the one established fact of Cloud technology is its
> nebulousness.

Keep those heads firmly in the sand.

[Stephen Hoffman]

On 2015-03-23 13:42:10 +0000, Bob Koehler said:

> In article <cnahfk...@mid.individual.net>,
> bi...@server3.cs.scranton.edu (Bill Gunshannon) writes:
>>
>> Might want to talk with him again. In the past week I have seen
>> numerous commercials lately claiming Accuweather is based on "The
>> Microsoft Cloud". I doubt that is VMS based.
>
> I suspect they have PCs on desks. And I suspect that is not thier
> compute engine for weather forcasting.

Particularly given the scale of computing involved with weather
forecasting, a back-end engine, Apollo <http://hp.com/go/apollo> would
probably be a better bet. If the folks are not going for something yet
bigger and faster. "Clouding" those forecasting models will likely be
more expensive than in-house computing, even given the need for
swapping out an Apollo-like box every few years.

> But VMS can play in an MS authentication and access MS servers, so I
> suspwect it could be convinced to access either of those "in the Cloud"
> since the one established fact of Cloud technology is its nebulousness.

Have you tried those OpenVMS interfaces, for this or for similar cases?
OpenVMS has no SMB/CIFS client short of command-line file transfers,
no WebDAV client short of command-line transfers, no distributed
computing support short of porting over or writing your own, and where
the external authentication is limited to account status and password
support whether connecting to Microsoft Active Directory, to Open
Directory or other LDAP-compatible servers. Could you connect OpenVMS
to "the cloud"? Sure. But there are easier ways to do that.


--
Pure Personal Opinion | HoffmanLabs LLC

[clairg...@gmail.com]

Accuweather is one of our Field test sites.

[David Froble]

clairg...@gmail.com wrote:
> Accuweather is one of our Field test sites.

One of the problems at DEC, which you're probably much more familiar
with than I am, was hearing that some VMS customers did not want to be
known. At least that is what some of us were told.

I think one thing that VSI might want to consider is to urge customers
to allow their names to be used in marketing. It is in their best
interest, I'd think, to have VSI be successful, and marketing to new
customers will to some extent determine that success.

Just my thoughts on the subject.

[David Froble]

Bill Gunshannon wrote:
> In article <XApkMI...@eisner.encompasserve.org>,
> koe...@eisner.nospam.decuserve.org (Bob Koehler) writes:
>> In article <cnahfk...@mid.individual.net>, bi...@server3.cs.scranton.edu (Bill Gunshannon) writes:
>>> Might want to talk with him again. In the past week I have seen numerous
>>> commercials lately claiming Accuweather is based on "The Microsoft Cloud".
>>> I doubt that is VMS based.
>> I suspect they have PCs on desks. And I suspect that is not thier
>> compute engine for weather forcasting.
>
> "The Microsoft Cloud" is not "PCs on desks". This whole "cloud" thing
> is being pushed as the ability to have massive compute power without
> having to buy the compute engines themselves.
>
>> But VMS can play in an MS authentication and access MS servers, so
>> I suspwect it could be convinced to access either of those "in the
>> Cloud" since the one established fact of Cloud technology is its
>> nebulousness.
>
> Keep those heads firmly in the sand.
>
> bill
>

Got to love that sand. At least as much as Accuweather being a VMS
field test site ....

:-)

You need to quit reverting to pre-VSI form ....

[clairg...@gmail.com]

On Monday, March 23, 2015 at 10:06:53 PM UTC-4, David Froble wrote:
> clairg...@gmail.com wrote:
> > Accuweather is one of our Field test sites.
>
> One of the problems at DEC, which you're probably much more familiar
> with than I am, was hearing that some VMS customers did not want to be
> known. At least that is what some of us were told.
>

This is true.

> I think one thing that VSI might want to consider is to urge customers
> to allow their names to be used in marketing. It is in their best
> interest, I'd think, to have VSI be successful, and marketing to new
> customers will to some extent determine that success.
>
> Just my thoughts on the subject.

We agree.

[VAXman-]

In article <dc2f7dde-a75d-4cc9...@googlegroups.com>, clairg...@gmail.com writes:
>Accuweather is one of our Field test sites.

Well then, that might squelch the M$ conjectures. ;)

[VAXman-]

In article <50881470-1dc3-481c...@googlegroups.com>, clairg...@gmail.com writes:
>On Monday, March 23, 2015 at 10:06:53 PM UTC-4, David Froble wrote:

>> clairg...@gmail.com wrote:
>> > Accuweather is one of our Field test sites.
>>
>> One of the problems at DEC, which you're probably much more familiar
>> with than I am, was hearing that some VMS customers did not want to be
>> known. At least that is what some of us were told.
>>
>

>This is true.

Albeit, Accuweather was featured in an HP OpenVMS video, so it's obvious that
Accuweather wasn't hiding behind any veil of customer secrecy.

[Bill Gunshannon]

In article <meqgq9$a1i$1...@dont-email.me>,

Yeah, lets see someone convince Accuweather to drop the "Microsoft
Cloud" commercials and publicly announce that they are running on VMS.
:-)

[Bill Gunshannon]

In article <00AF498D...@sendspamhere.org>,

VAXman- @SendSpamHere.ORG writes:
> In article <50881470-1dc3-481c...@googlegroups.com>, clairg...@gmail.com writes:
>>On Monday, March 23, 2015 at 10:06:53 PM UTC-4, David Froble wrote:
>>> clairg...@gmail.com wrote:
>>> > Accuweather is one of our Field test sites.
>>>
>>> One of the problems at DEC, which you're probably much more familiar
>>> with than I am, was hearing that some VMS customers did not want to be
>>> known. At least that is what some of us were told.
>>>
>>
>>This is true.
>
> Albeit, Accuweather was featured in an HP OpenVMS video, so it's obvious that
> Accuweather wasn't hiding behind any veil of customer secrecy.

Apparently they are now. Or is the "Microsoft Cloud" really a bunch of
VMS machines? Am I the only one who sees this as a problem? How do you
sell a product when someone with large visibilty that you claim is a
major VMS site publicly states, quite to the contrary, that they are a
Microsoft Cloud company?

I'm not knocking VMS. I am nerely pointing out that this is a serious
piece of anti-advertising.

[Bill Gunshannon]

In article <00AF498C...@sendspamhere.org>,

VAXman- @SendSpamHere.ORG writes:
> In article <dc2f7dde-a75d-4cc9...@googlegroups.com>, clairg...@gmail.com writes:
>>Accuweather is one of our Field test sites.
>
> Well then, that might squelch the M$ conjectures. ;)
>

Not when the only public presence provided by Accuweather themselves
says they are "Microsoft Cloud" and has no mention of VMS at all.
The commercials I saw were all placed in programs where they are
likely to be seen by the people with the buying power and corporate
influence in places that VMS should be fighting for as customers.
This isn't going to help.

[johnwa...@yahoo.co.uk]

What's not going to help is the belief that any organisation shall
worship at one (and exactly one) IT altar.

It is actually possible for an organisation to have more than one piece
of an IT strategy, and be happy with all of it. One size does not fit all.

When the provider of the most expensive bit of the organisation's IT says
"there's (e.g.) 30% off next year if you let us use you as a reference site",
how many organisations will refuse?

30% off many organisation's Microsoft bill is a lot of pennies.

30% off many organisation's VMS bill is down in the noise.

[Bill Gunshannon]

In article <a61e3181-aad4-46d0...@googlegroups.com>,

And, while all of that is correct, what do you think the influence will
be when the CIO of a major corporation is left with the choice of VMS
or Microsoft Cloud? Microsoft is already on all their desktops, just
why should he even consider bringing VMS in? One size does not always
fit all, but the door is seldom open for the odd man out. He has to
fight harder to get in that door.

[Kerry Main]

> -----Original Message-----
> From: Info-vax [mailto:info-vax...@info-vax.com] On Behalf Of

> Bill Gunshannon
> Sent: 24-Mar-15 7:57 AM
> To: info...@info-vax.com
> Subject: Re: [New Info-vax] Accuweather new contract
>
> In article <meqgq9$a1i$1...@dont-email.me>,
> David Froble <da...@tsoft-inc.com> writes:
> > clairg...@gmail.com wrote:
> >> Accuweather is one of our Field test sites.
> >
> > One of the problems at DEC, which you're probably much more familiar
> > with than I am, was hearing that some VMS customers did not want to
> be
> > known. At least that is what some of us were told.
> >
> > I think one thing that VSI might want to consider is to urge customers
> > to allow their names to be used in marketing. It is in their best
> > interest, I'd think, to have VSI be successful, and marketing to new
> > customers will to some extent determine that success.
> >
> > Just my thoughts on the subject.
>
> Yeah, lets see someone convince Accuweather to drop the "Microsoft
> Cloud" commercials and publicly announce that they are running on
> VMS.
> :-)
>

If a Cust is using Office365, or their SharePoint cloud offering, then MS
could legitimately claim Cust X is using the Microsoft Cloud. They may
have some Azure based applications which would also be legitimate
claims to be using the Microsoft cloud.

That would not change the fact that OpenVMS is being used in other
areas of the company.

One needs to understand that the underlying platform is seldom of
interest to the senior execs in any Cust environment. It is usually
only of interest to the IT folks and the vendor of that platform.

[David Froble]

Microsoft desktop apps do the jobs needed on the desktops. VMS doesn't
have the desktop apps. It's reasonable to use what is needed for the
job. If a CIO doesn't understand that, I've got to wonder who hired the
idiot.

But there are other needs, for which perhaps the desktop apps are not so
well suited. A decent CIO does understand this also. For some of those
needs, VMS is a viable solution. A decent CIO can also understand this.

There are what we're calling desktop applications in just about every
organization. Always have been. In the past they were manual, and then
mechanical, and then users on a computer system, and then when PCs
became available on individual desktop systems. There was a huge
unfulfilled need for what became desktop systems, and when such finally
became available, that need far over shadowed other computer usage, and
swept Microsoft to where it is today.

However, the desktop systems have never been capable of doing some of
the other jobs for which computers have been used. Some have attempted
to do so. Probably many still survive, but I'm aware of some that
haven't survived their poor judgment.

Getting back to Accuweather. They distribute weather information. For
many, that distribution is to desktop, notebook, tablet, and smart phone
systems. Probably the majority to smart phones today. But unless
they're just re-routing weather information provided by others, they
might need some serious computing capability. Probably not provided by
a bunch of smart phones.

[Bill Gunshannon]

In article <mailman.1.1427203772.92...@info-vax.com>,

Kerry Main <kerry...@backtothefutureit.com> writes:
>> -----Original Message-----
>> From: Info-vax [mailto:info-vax...@info-vax.com] On Behalf Of
>> Bill Gunshannon
>> Sent: 24-Mar-15 7:57 AM
>> To: info...@info-vax.com
>> Subject: Re: [New Info-vax] Accuweather new contract
>>=20
>> In article <meqgq9$a1i$1...@dont-email.me>,
>> David Froble <da...@tsoft-inc.com> writes:
>> > clairg...@gmail.com wrote:
>> >> Accuweather is one of our Field test sites.
>> >
>> > One of the problems at DEC, which you're probably much more familiar
>> > with than I am, was hearing that some VMS customers did not want to
>> be
>> > known. At least that is what some of us were told.
>> >
>> > I think one thing that VSI might want to consider is to urge customers
>> > to allow their names to be used in marketing. It is in their best
>> > interest, I'd think, to have VSI be successful, and marketing to new
>> > customers will to some extent determine that success.
>> >
>> > Just my thoughts on the subject.

>>=20

>> Yeah, lets see someone convince Accuweather to drop the "Microsoft
>> Cloud" commercials and publicly announce that they are running on
>> VMS.
>> :-)

>>=20
>
> If a Cust is using Office365, or their SharePoint cloud offering, then MS=20
> could legitimately claim Cust X is using the Microsoft Cloud. They may=20

> have some Azure based applications which would also be legitimate
> claims to be using the Microsoft cloud.

Well, the pictures shown in the commercial were not people using Word
or Excel. They were people doing weather modeling.

>
> That would not change the fact that OpenVMS is being used in other=20
> areas of the company.

And yet, we never see this in a TV commercial. Perception is reality.

>
> One needs to understand that the underlying platform is seldom of
> interest to the senior execs in any Cust environment. It is usually
> only of interest to the IT folks and the vendor of that platform.

And they are not the ones who sign the purchase orders. How can one
expect the CIO to sign a PO for a product he has never heard of? He
won't hesitate to sign one for Microsoft Cloud Services.

[Bill Gunshannon]

In article <mersp6$m27$1...@dont-email.me>,

You really don't get out much, do you. CIO's see things like that
Microsoft Cloud commercial and that strongly stters their decisions
in that direction. And this is nothing new. I mentioned having a
DECStation when I first came here. When I needed it replaced with
something more powerful (and more up to date as the DECStation ran
the same Ultrix the VAXStation did!!) I had speced out a Sun based
on capabilities and cost. My boss approved it. The next level
manager who actually signed the checks came back from an executive
level (as opposed to techie level) conference where HP had demoed
HPUX workstations. He was sold. No Sun, had to get an HP. After
all, the salesdroid had put on a real impressive dog and pony show.
When we got the workstaion at about 50% higher cost it could not do
any of the stuff the salesdroid had demoed because his system was
fully loaded and ours was not. Neither cold it do most of the
stuff it was purchased for. It ended out being used a the University's
DNS Server. Nice job for over $20K. The point being, the man who
makes the decision is not a technical expert he is a manager. He
makes his decision from management data. And he buys Microsoft Cloud.

>
> But there are other needs, for which perhaps the desktop apps are not so
> well suited. A decent CIO does understand this also. For some of those
> needs, VMS is a viable solution. A decent CIO can also understand this.

See above. CIO's are managers, not techies. they may have been techies
at some point in their careers but that is not what got them into that
corner office. An MBA did.

>
> There are what we're calling desktop applications in just about every
> organization. Always have been. In the past they were manual, and then
> mechanical, and then users on a computer system, and then when PCs
> became available on individual desktop systems. There was a huge
> unfulfilled need for what became desktop systems, and when such finally
> became available, that need far over shadowed other computer usage, and
> swept Microsoft to where it is today.

see my comment to Kerry. The commercial doesn't show Accuweather
employess using word or Excel. It shows them using weather modeling
software and it says they are doing it on The Microsoft Cloud. What's
more, the background comments are things like "We couldn't do this
without The Microsoft Cloud". Doesn't say much about their opinion
of VMS, does it. And this is supposed to be one of VMS' most important
customers? With friends like this........

>
> However, the desktop systems have never been capable of doing some of
> the other jobs for which computers have been used. Some have attempted
> to do so. Probably many still survive, but I'm aware of some that
> haven't survived their poor judgment.

Get off this "desktop" kick. That is not what the commercial was
selling. And that is certainly not what anyone means when they say
"The Microsoft Cloud".

>
> Getting back to Accuweather. They distribute weather information. For
> many, that distribution is to desktop, notebook, tablet, and smart phone
> systems. Probably the majority to smart phones today. But unless
> they're just re-routing weather information provided by others, they
> might need some serious computing capability. Probably not provided by
> a bunch of smart phones.

Get out more. Maybe the commercial is available on YouTube or something.
Look at what they are actually saying. And then remember who they are
targetting this message at. It really makes little difference what
VSi does if the people making the decisions are still being steered
away from it.

[JF Mezei]

The best way for VSI to get some exposure is to get friendly with the
main tech reporters and make themselves available whenever needed.

Remember that reporters are impatient, and if they don't get a call back
in 30 seconds you get mentioned as "was not available" or "did not
return calls"

Once reporters know you exist, you then get your customers to just
mention they use VMS. Reporters can then contact you for information
about state of VMS.

and of course, VSI has to learn how to do press releases. (And this is
where Twitter gets interesting because issuance of press release gets
wider audience than just those looking for it).

[mcle...@gmail.com]

On Wednesday, March 25, 2015 at 3:32:18 AM UTC+11, Bill Gunshannon wrote:
> In article <mersp6$m27$1...@dont-email.me>,

In my experience PHB's seem to decide to purchase or use whatever's getting lots of mentions in the IT press. Hardware, software, software development methodology .. it doesn't seem to matter. Maybe this is an extension of the old "No-one gets fired for buying IBM", but this time it's "No-one gets fired for doing what seems popular".

Lots of mention of the "cloud". Almost no mention of VMS for 20 years. Which do you think will be popular and which be very rare?

[David Froble]

He should. He really should ....

[Kerry Main]

> -----Original Message-----
> From: Info-vax [mailto:info-vax...@info-vax.com] On Behalf Of

> mcle...@gmail.com
> Sent: 24-Mar-15 5:17 PM
> To: info...@info-vax.com
> Subject: Re: [New Info-vax] Accuweather new contract
>

[snip...]

>
> In my experience PHB's seem to decide to purchase or use whatever's
> getting lots of mentions in the IT press. Hardware, software, software
> development methodology .. it doesn't seem to matter. Maybe this is an
> extension of the old "No-one gets fired for buying IBM", but this time it's
> "No-one gets fired for doing what seems popular".
>
> Lots of mention of the "cloud". Almost no mention of VMS for 20 years.
> Which do you think will be popular and which be very rare?

Let's remember that terms like SOA were also bandied around for a
number of years by the media and vendors as THE solution all Custs
had to have. SOA washing (every product would help SOA somehow)
became the norm.

Many Cust's jumped on the SOA band wagon and ultimately many
got burned big time. This is not to say that SOA does not have some
good points, but it got misused and mis-positioned, so the result was
many projects that went south big time. Now, many senior Custs will
throw you out on your ear if you bring up a SOA discussion with them.

The concept of distributed services has been around since the DEC
Days - anyone remember NAS (network app services)?

This is where "cloud" computing is today. It has been so misused
that I expect a Cust back lash similar to SOA.

Minus the hype -

Public clouds = selective IT outsourcing

Private clouds = internal shared services

Pros and cons with each strategy, but these are no different than
what has been around for decades.

Given all of the high profile security challenges, concerns & outages
with public clouds, there has been reports of something like 80%
of Cust's looking at cloud are looking at private clouds. Yes, there are
exceptions, but not the majority.

[David Froble]

mcle...@gmail.com wrote:

> Lots of mention of the "cloud". Almost no mention of VMS for 20
> years. Which do you think will be popular and which be very rare?

The freeking cloud is just timesharing all over again, with much better
communications.

Just wondering when some marketing type will be bringing back the hula
hoop ???

[Kerry Main]

> -----Original Message-----
> From: Info-vax [mailto:info-vax...@info-vax.com] On Behalf Of

> David Froble
> Sent: 24-Mar-15 9:36 PM
> To: info...@info-vax.com

Opportunity knocks .. how about a new invention called a "cloud" ring?

:-)

[David Froble]

Kerry Main wrote:
>> -----Original Message-----
>> From: Info-vax [mailto:info-vax...@info-vax.com] On Behalf Of
>> David Froble
>> Sent: 24-Mar-15 9:36 PM
>> To: info...@info-vax.com
>> Subject: Re: [New Info-vax] Accuweather new contract
>>
>> mcle...@gmail.com wrote:
>>
>>> Lots of mention of the "cloud". Almost no mention of VMS for 20
>>> years. Which do you think will be popular and which be very rare?
>> The freeking cloud is just timesharing all over again, with much better
>> communications.
>>
>> Just wondering when some marketing type will be bringing back the hula
>> hoop ???
>
> Opportunity knocks .. how about a new invention called a "cloud" ring?
>
> :-)

No, really, I'm serious about this.

Back in the day, the IT department provided services for the users.
Sometimes those services were perceived to be inadequate, or slow to
materialize, and such, and is one reason the PCs spread so quickly.
Users could afford them, and didn't have to wait for whatever IT would
dole out to them.

So, what is a private cloud? It's the centralization of IT, just like
back in the day.

The company I worked for started out selling timesharing services on a
RSTS system, with 110 BAUD accoustical modems. Other than much better
communications today, what's so different about a public cloud?

It used to be the cost of HW that kept things centralized. Today it's
maybe the cost of people, power, office space, whatever.

Private centralization of services I have no problem with. As for the
public services, maybe useful for some things, but to me it's like
putting all your eggs in somebody else's basket. Life on the edge ..

[IanD]

...and the fact that the organisation itself doesn't have to fork out the initial cost for the hardware and can have it provisioned in a fraction of the time of ordering in a new system and getting it set up and configured. The other advantage is at tear down time, the cost stop. Bean counters love it

I'd term it as 'time-sharing for the short-term focused' :-)

I'm only a partial advocate for cloud, having played with Amazon's offering and seeing how easy it is to spin up an instance, it's pretty dam impressive

I'd really recommend for VMS folk to sign up for an Amazon free tier instance and then go through the process of getting a system up and running - it's an eye opener if you have never done this before to see where VMS needs to be in terms of infrastructure readiness and ease of systems infrastructure provisioning - we have a long way to go

There's a good introduction to Amazons AWS at cbt nuggets, just smile with your teeth firmly clenched at the presenters over enthusiastic demeanour about all things AWS and you'll get through it without barfing :-)

As to Accuweather not using the VMS naming anywhere, perhaps this was deliberate ? I know I'm going to probably get frowned upon for saying this but perhaps they didn't want to advertise the fact that they are using VMS because of a perceived negative connotation with it being old and antiquated and not modern?

<Searches for bunker to hide in after dropping a VMS negative...>

[Kerry Main]

It's what is typically called internal shared services. Rather than have
many BU's provide their own non-std IT services, the strategy states
that all common tactical IT services will be delivered by a single BU
(central IT) with strategic services like App dev/support continuing
to be supplied by the BU's / depts. themselves.

Universities are famous for wild west IT environments whereby each
College / dept does their own thing. (insert Bills views here)

Reference:
http://tinyurl.com/Cambridge-DC-Consolidation
" In this Q&A, learn how an 800-year-old institution plans to put 200
server rooms from over 100 self-steering departments under one roof
and one management strategy."

> The company I worked for started out selling timesharing services on a
> RSTS system, with 110 BAUD accoustical modems. Other than much
> better
> communications today, what's so different about a public cloud?
>

Public cloud is simply selective IT outsourcing. Like any other form of
outsourcing there are pro's and con's.

> It used to be the cost of HW that kept things centralized. Today it's
> maybe the cost of people, power, office space, whatever.
>

Generally speaking:
Centralized = Good resource management and cost controls. Good DR
features and capabilities.
Centralized = Bad customer service and poor adaptability to rapidly
changing business requirements.

Decentralized = Good Customer Service, good end user communications
and good adaptability to changing business requirements.
Decentralized = Very bad cost controls, lack of standardization, poor
resource management, poor DR capabilities.

Hence, the answer is neither centralized or decentralized. The real
answer is to find a shared services model which uses the best features
of both centralized and distributed models to match your companies
specific requirements.

> Private centralization of services I have no problem with. As for the
> public services, maybe useful for some things, but to me it's like
> putting all your eggs in somebody else's basket. Life on the edge ..

Today's IT challenges are no different than they were decades ago.
Yes, new technologies, but the best practices from an IT perspective
are timeless and is the reason why "dinosaurs" (read experienced)
are constantly experiencing deja vue.

[Kerry Main]

> -----Original Message-----
> From: Info-vax [mailto:info-vax...@info-vax.com] On Behalf Of
> IanD
> Sent: 24-Mar-15 10:45 PM
> To: info...@info-vax.com
> Subject: Re: [New Info-vax] Accuweather new contract
>

> On Wednesday, March 25, 2015 at 12:30:00 PM UTC+11, David Froble
> wrote:
> > mcle...@gmail.com wrote:
> >
> > > Lots of mention of the "cloud". Almost no mention of VMS for 20
> > > years. Which do you think will be popular and which be very rare?
> >
> > The freeking cloud is just timesharing all over again, with much better
> > communications.
> >
> > Just wondering when some marketing type will be bringing back the
> hula
> > hoop ???
>
> ...and the fact that the organisation itself doesn't have to fork out the
> initial cost for the hardware and can have it provisioned in a fraction of
> the time of ordering in a new system and getting it set up and
> configured. The other advantage is at tear down time, the cost stop.
> Bean counters love it
>
> I'd term it as 'time-sharing for the short-term focused' :-)
>
> I'm only a partial advocate for cloud, having played with Amazon's
> offering and seeing how easy it is to spin up an instance, it's pretty dam
> impressive
>

The same can be done with internal shared services without all of the
security concerns many cust's have with putting their data on the
Internet with minimal influence of what happens when their services
are unavailable (call the 800 line and line up with your other million
closest friends).

While developer types love the ability to "spin things up" at will,
Company execs are not so easily impressed. They worry about
unplanned budget hits, VM sprawl as well as things like data backups
and off site archiving. (Hint, many cloud providers offer replication,
but not backups and/or offsite archives. Some companies require 7+
years to be archived.

Best description of public cloud computing I read recently described
Public computing being like computing in a stall in a public washroom
i.e. you never know who is in the next stall.

:-)

> I'd really recommend for VMS folk to sign up for an Amazon free tier
> instance and then go through the process of getting a system up and
> running - it's an eye opener if you have never done this before to see
> where VMS needs to be in terms of infrastructure readiness and ease of
> systems infrastructure provisioning - we have a long way to go
>

Internal shared services can do the same thing today. Many Cust's have
used VMware type technologies to implement similar services.

> There's a good introduction to Amazons AWS at cbt nuggets, just smile
> with your teeth firmly clenched at the presenters over enthusiastic
> demeanour about all things AWS and you'll get through it without barfing
> :-)
>
> As to Accuweather not using the VMS naming anywhere, perhaps this
> was deliberate ? I know I'm going to probably get frowned upon for
> saying this but perhaps they didn't want to advertise the fact that they
> are using VMS because of a perceived negative connotation with it being
> old and antiquated and not modern?
>

When was the last time you bragged to your friends about how good
the services of your Telco are (dial tone always available)?

> <Searches for bunker to hide in after dropping a VMS negative...>

Not a problem - OpenVMS has its challenges and no one here says
It does not need improvements.

However, Folks really need to better understand how big companies
work and how little their senior non-IT execs care about what OS
platform their app services run on. They care about SLA's being met
and whether that App service provides them with a competitive
advantage.

As the OpenVMS testimonials posted earlier for Accuweather show, the
App services running on OpenVMS meets and exceeds their SLA's, it also
provides them with a competitive advantage.

[Stephen Hoffman]

On 2015-03-25 02:29:05 +0000, David Froble said:

> No, really, I'm serious about this.

HP and IBM are aimed right at this market, as are various other vendors
— such as some Adobe products — to various degrees.

HP's big push here is with Helion <http://hp.com/go/helion> and (of
course) HP hardware, running with either HP servers or with your own
local servers, or both.

I do wish there was less jargon swirling here in general, as there are
some very useful products available now — but that can get obscured in
the blizzards of marketing.



--
Pure Personal Opinion | HoffmanLabs LLC

[Bob Koehler]

In article <0e8f8a75-76d3-4bad...@googlegroups.com>, mcle...@gmail.com writes:
>
> Lots of mention of the "cloud". Almost no mention of VMS for 20 years. Whi=

> ch do you think will be popular and which be very rare?

So OK, stealth marketing helped kill DEC. Perhaps you could believe
that some VSI folks are painfully aware of this?

No reason VMS couldn't participate in "Cloud". But I think VSI has
to finish the 86 port before they invest further in VMS' ability
to do so.

[Bob Koehler]

In article <met63o$5f5$1...@dont-email.me>, David Froble <da...@tsoft-inc.com> writes:
>
> So, what is a private cloud? It's the centralization of IT, just like
> back in the day.

Nope. "Cloud" is whtever the vendor says it is. It's about as well
defined as the number of angles that can dance on the head of a pin.

[mcle...@gmail.com]

On Thursday, March 26, 2015 at 5:02:59 AM UTC+11, Bob Koehler wrote:

> In article <met63o$5f5$1...@dont-email.me>, David Froble <dav...@tsoft-inc.com> writes:
> >
> > So, what is a private cloud? It's the centralization of IT, just like
> > back in the day.
>
> Nope. "Cloud" is whtever the vendor says it is. It's about as well
> defined as the number of angles that can dance on the head of a pin.

Angles? Surely angels.

Someone here once described the cloud as mistware or fogware...

[VAXman-]

In article <cnd1vc...@mid.individual.net>, bi...@server3.cs.scranton.edu (Bill Gunshannon) writes:
>In article <00AF498D...@sendspamhere.org>,
> VAXman- @SendSpamHere.ORG writes:

>> In article <50881470-1dc3-481c...@googlegroups.com>, clairg...@gmail.com writes:
>>>On Monday, March 23, 2015 at 10:06:53 PM UTC-4, David Froble wrote:
>>>> clairg...@gmail.com wrote:
>>>> > Accuweather is one of our Field test sites.
>>>>
>>>> One of the problems at DEC, which you're probably much more familiar
>>>> with than I am, was hearing that some VMS customers did not want to be
>>>> known. At least that is what some of us were told.
>>>>
>>>
>>>This is true.
>>
>> Albeit, Accuweather was featured in an HP OpenVMS video, so it's obvious that
>> Accuweather wasn't hiding behind any veil of customer secrecy.
>

>Apparently they are now. Or is the "Microsoft Cloud" really a bunch of
>VMS machines? Am I the only one who sees this as a problem? How do you
>sell a product when someone with large visibilty that you claim is a
>major VMS site publicly states, quite to the contrary, that they are a
>Microsoft Cloud company?
>
>I'm not knocking VMS. I am nerely pointing out that this is a serious
>piece of anti-advertising.

I've never seen an Accuweather commercial.

[MICHAEL W FARRELL]

I was not aware of Accuweather until about two weeks ago, when Verizon FIOS,
my cable company (permanently) dropped the Weather Channel and replaced it
with Accuweather.!!

> _______________________________________________
> Info-vax mailing list
> Info...@rbnsn.com
> http://rbnsn.com/mailman/listinfo/info-vax_rbnsn.com

[Bob Koehler]

In article <b430fadc-0871-42ba...@googlegroups.com>, mcle...@gmail.com writes:
> On Thursday, March 26, 2015 at 5:02:59 AM UTC+11, Bob Koehler wrote:
>> In article <met63o$5f5$1...@dont-email.me>, David Froble <dav...@tsoft-inc.com> writes:
>> >
>> > So, what is a private cloud? It's the centralization of IT, just like
>> > back in the day.
>>
>> Nope. "Cloud" is whtever the vendor says it is. It's about as well
>> defined as the number of angles that can dance on the head of a pin.
>
> Angles? Surely angels.

Freudian slip. Coudn't define either one.

[Jess Goodman]

VAXman wrote:
"Jess Goodman would be a good candidate to comment on this. I'm pretty sure he lurks and reads c.o.v."

I have very little lurking time available now-a-days, but eventually I did stumble upon this thread. My bad luck since I guess I am now forced to comment upon it.

My apologies in advance, but because this thread is in an open forum I will not be able to address all of the questions and speculations about AccuWeather contained in it. Nor can I be very specific in what I am allowed to say. But with that said, I can say this:

o I am employed at AccuWeather as the OpenVMS Administrator and Senior Engineer for VMS development.
o AccuWeather also employs Administrators and developers for other OSs, including Microsoft Windows.
o The recent TV commercials about AccuWeather were made by and paid for by Microsoft.
o The not-so-recent (2004) HP success stories about AccuWeather were made by and paid for by HP.
o Both sets of ads featured AccuWeather in a very favorable light. Of course we were delighted to participate in them, regardless of what OS they might be promoting.
o The only OSs that AccuWeather has ever deliberately promoted are, due to our leadership position in mobile weather apps, Android and IOS.
o We have never hidden the fact that AccuWeather uses VMS, although the Microsoft video crew did so (literally) while they were filming here.
o We have fewer VMS developers (and many more Windows developers) now than we had in 2004.
o Our meteorologists still log into our VMS cluster (or, as they insist on calling it, "the VAX") from PCs using PowerTerm, but they now use a browser and our internal web site a lot more than they use a DCL command line.
o A significant part of our back-end forecasting services still runs under VMS (Alpha and Itanium), but in 2004 it almost all ran on VMS, as did the back-end services for our web site.

Over 10 years ago, just after HP released AccuWeather's "success" story, there was a C.O.V. thread very similar to this one. I was not allowed to inject any context to HP's story about us back then, but I can say now what I was thinking then. From that story:

<< With Attunity Connect, the AccuWeather website seamlessly accesses weather conditions and forecast data directly from the OpenVMS systems, eliminating the bottlenecks and performance problems that were introduced when the data had to be replicated in SQL Server.

"We decided to stay on OpenVMS because the platform still has a lot of life in it. The Attunity solution allowed us to remain on OpenVMS and leverage its power, speed, accuracy and security," states Fiore. >>

What Kathy Fiore, our CIO at that time, said then was entirely true, but take a moment and try to read between the lines. I'll wait... OK, now - do you still think that this story should go in the win column for VMS? I don't!

"We decided to stay on OpenVMS...". In other words we had almost decided to move all of the back-end servicing of our public web site off of VMS.

"...the platform still has a lot of life in it." In other words it may be dying, but it 'aint dead yet.

"With Attunity Connect the website seamlessly accesses...data directly from OpenVMS..." Attunity Connect is excellent middleware, but it is still middleware, and IMO middleware does not belong in any "final" design. It was a huge help to us at first, but after a while it turned into something that was getting in our way more than it was helping us. So eventually the web-site back-end was moved off of VMS anyway.

"...bottlenecks and performance problems that were introduced when the data had to be replicated..." The obvious but unanswered question is: Why did this data have to be replicated in the first place? The answer: because there was no webserver for VMS that could scale to anywhere near the performance capacity that AccuWeather required.

J. Goodman
"I have one but it's personal."

[David Froble]

Jess Goodman wrote:

> "...bottlenecks and performance problems that were introduced when
> the data had to be replicated..." The obvious but unanswered
> question is: Why did this data have to be replicated in the first
> place? The answer: because there was no webserver for VMS that could
> scale to anywhere near the performance capacity that AccuWeather
> required.

I see this as a problem resulting from the absurd idea of "free
software". Once someone gets the idea of such, then the next step is
"the price is right, make it work", and now you're limited to whatever
someone will choose to implement, without pay, on their own time.

Needless to say, that "free application" just might not be available in
every environment. Or when ported to another environment, might not
work so well.

I'd also wonder at just what "performance capacity" is required, or
currently in use at for instance AccuWeather? Why would the WASD web
server not satisfy that demand? Not saying it should, just asking where
it might not.

I don't use a web server. One of my shortcomings, perhaps, perhaps not.
To me a web server is just a listener socket, accepting connection
requests, with some capability to do something with some of the data
sent by a client. Perhaps over-simplified, but, that's what it does.

With sufficient specifications, I cannot see why a web server could not
be implemented on VMS to satisfy any requirements. Even a large cluster
running multiple copies of the web server to meet the volume of
connection requests. (No, I haven't volunteered to write such.)

Also, you cannot solve a problem, unless you know what the problem is.
I'd be curious as to where and why VMS based web servers do not satisfy
the requirements?

The subject interests me because a cousin of the web server, a web
service, has greatly influenced the applications he's working with.

In the old days, people would sit at terminals, with phones, and takes
customer orders over the phone. (Dave shows his extreme age.) We have
implemented "web services" and a protocol to enable customers to submit
orders computer-to-computer. Again, basically a listener socket, with
code to process specific requests from clients. We provide inventory
inquiry, process sales orders, and such. The providers of software on
the other end are loving it, and have taken steps to use this
capability. It actually seems to be a bit astounding, how much this has
changed the Codis customer's business.

Slightly off topic story.

We've been discussing how to set up things with more disaster tolerance.
There have been few problems in the past, but, things can always be
made better. Now, I was a bit confused with the "sudden" interest in
disaster tolerance. The question was, "what happens if we lose a day's
work (orders)?" I wondered what the problem was. After all, we're
selling a few lawn mower parts, how much could be lost? A couple
thousand dollars worth of orders?

Yesterday I brought up that very question. The answer. "Oh, no, the
customers have been expanding, a million dollar day is possible ...."

.
.
.

(Dave hits floor ....)

.
.
.

While laying on floor, stunned, Dave considers his (possibly out of
date) labor rates ....

Further discussion discloses that financial reporting from the GL system
has had to be modified to handle the greatly increased figures ....

(More thoughts on out of date labor rates ....)

Ok, enough on the story ..

But one thing is clear. The customers have increased business greatly,
and part of that is because of the new web services that we've
developed. It has been a real "changer" of the business.

As a simple example. The dealers now have small systems, and usually
have software that provides a "parts explosion" of the products they are
working on. If they select a particular product, the dealer system can
obtain a real time inventory availability on every distributor running
Codis, and let the dealer know what's available. The dealer system can
then take the selected part(s) and immediately send an order to the
distributor, who can fill the order from inventory, or, another
distributor using Codis. Do that on the phone ....

So, now, I can say that I understand just how important it is to embrace
and develop some new ways of doing business. We, and I assume others,
have been able to do it on VMS. The question then becomes, why are
others having problems doing so, and what needs to happen to rectify
that situation?

I can specify one thing, the disgusting TCP/IP from HP. I'm glad VSI is
going to look at this.

[johnwa...@yahoo.co.uk]

Dave said "whatever someone will choose to implement, without pay, on
their own time. "

Dave, with the greatest respect, you may need to get out more :)

Various Linuxes may be free to download but there are lots of people
being paid by their big-name employers to do Linux development. The
proportion of input from unpaid volunteers isn't quite negligible but
it seems it's a lot smaller than you've been thinking (how does 13%
sound?).

The Linux Foundation does a survey of where the Linux changes come
from, and sometimes it gets reported. Apparently you'd be surprised by
the numbers. Perhaps others would too. So here's a recent sample.

E.g. in 2013 [1], top corporate contributor was Red Hat at 10% of
all changes. Next big names, in order, were Intel (9%), TI (4%),
[Linaro (4%) aren't famous], Suse (3%), IBM (3%), Samsung (2%),
and Google (2%). There are plenty more corporate names on the list,
not all of them widely known in the land of VMS and Basic, but in
their fields they are big names: e.g. Oracle, Broadcom, Qualcomm,
Cisco, etc.

In 2013 the proportion of changes submitted by sole
developers working on their own was only 13%.

One well known name not showing on that list: HP. There is a category
for "unknown", at 3%.

I realise statistics are a wonderful thing, and are more often used
for support than for illumination. But have a think about this. Linux
isn't $$$ free, it's paid for by the customers of the companies whose
employees are paid to contribute.



Does this change anything in your picture? Not sure. But it might.


[1] http://arstechnica.com/information-technology/2013/09/google-and-samsung-soar-into-list-of-top-10-linux-contributors/
which in turn reflects
http://www.linuxfoundation.org/news-media/infographics/who-writes-linux-2013
Slightly better analysis for 2013's changes (but fewer raw numbers) at
http://www.infoworld.com/article/2610207/open-source-software/who-writes-linux--corporations--more-than-ever.html

For the equivalent info for changes during 2014 see
http://arstechnica.com/information-technology/2015/02/linux-has-2000-new-developers-and-gets-10000-patches-for-each-version/
but in summary it's not that different than the previous year and HP are
still invisible (? there's stuff like HP's involvement with OpenStack. Right.)

[JF Mezei]

On 15-03-28 15:46, johnwa...@yahoo.co.uk wrote:

> One well known name not showing on that list: HP. There is a category
> for "unknown", at 3%.

This in interesting in light of the fact that HP's enterprise strategy
is to move away from HP-UX to Linux/Windows.

On the other hand perhaps not surprising. HP's strategy may be to port
its enterprise "portfolio" of technologies to proprietary products on
Linux, so HP can monetize selling Linux with plenty of HP-only bells and
whistles.

This would result in HP making minimal open-source contribitions to
standard Linux, but providing many middleware/layered products that run
on Linux.

As well, consider that since HP laid off most of its OS engineering
staff, it has less ability to contribute. As well, its R&D arm, which
as severely cut under Hurd would have a hard time justifying budgets for
Linux cvontributions which have 0 chance (by design) of contributing to
HP's revenus.

The money folks like IBM spend to contribute to Linux have intangible
benefits, mostly good PR, as well as providing some steering of Linux to
better serve IBM's customers. But for this to count, you need a CEO with
vision that is not bean-counter driver with short term profit requirements.

[johnwa...@yahoo.co.uk]

JF wrote

"HP making minimal open-source contribitions to standard Linux, but
providing many middleware/layered products that run on Linux."

You might possibly think that. I couldn't possibly comment [1].

What I will say is that if Linux doesn't run (well) on Proliant, that
would presumably be a rather shortsighted move on HP's part. Unless
their goal is to shift all the server hardware sales from Proliant to
Foxconn/Cloudline.

[1] Not a comment, a question: What HP-unique middleware/LPs is there
(for x86-64) these days anyway? Some HP-UX derived stuff, maybe?

[JF Mezei]

On 15-03-28 16:24, johnwa...@yahoo.co.uk wrote:

> What I will say is that if Linux doesn't run (well) on Proliant, that
> would presumably be a rather shortsighted move on HP's part.

HP _could_ produce the patches and drivers specific to its server
designs and make them available separately, which means they wouldn't
been seen as a contribution to Linux itself since they wouldn't be
included in the generic distros. (I use "could* here to show on possible
way where HP's Linux spending/contreobutions would not be counted in
those statistics of contributors to Linux).

> [1] Not a comment, a question: What HP-unique middleware/LPs is there
> (for x86-64) these days anyway? Some HP-UX derived stuff, maybe?

At the start of the BCS wind down, HP had made statements that they
would be shifting much of the HP-UX enterprise software to Linux to make
it easier for customer to migrate. (this was early in Meg Whitman's
tenure). There was even a code name for this program "Project Odyssey"
I don't think anything came of it. But HP was using that "project" in
its financial teleconferences to provide confidence that HP was able to
bridge the gap between its legacy BCS that was going down fast and new
focus on industry standard servers with Windows-Linux as its new
enterprise strategy.


One aspect of HP's enterprise is that they are good at using buzzwords,
but apepar to hve difficulty in delivering specifics.

I do believe however that its system/network management software HP
Openview (it may have been renamed to "OneView") would support Linux,
but not sure if the management station runs on Linux).

[johnwa...@yahoo.co.uk]

Openview is a buzzword (a marketechture), it's not a specific product.

If you want to be very generous you could call it a family of products.

But see what you find if you look for Openview on a famous search engine.
The paid-for result (paid by HP?) says go to www.hp.com/OpenView
So, have a look there. Can you work out what are they offering that
customers can buy?

Back to Proliant - the Linux-specific stuff to make Proliant servers
work *well* with Linux used to be shipped with a thing called a
SmartStart CD which came with Proliant servers and support contracts.
No idea what the current state of play is.

In passing: for unrelated reasons, as I'm tryping this I'm also
installing an old SuSe64 (11.0) on an old Proliant DL360 g4. It works
out of the box, but there is no visible HPQ added value. Then again, I
didn't bother with the SmartStart CD - why would I, if the HPQ input
had been incorporated into the default Linux stuff. From the HPQ
website it *looks* as though the first HPQ-suppored DL360 for SUSE
Linux is the DL360 g5. So maybe I shouldn't be trying this with the
g4/11.0 combo anyway.


As the sUsE folks like to say: Have a lot of fun.

[Jan-Erik Soderholm]

David Froble skrev den 2015-03-28 19:52:
> Jess Goodman wrote:
>
>> "...bottlenecks and performance problems that were introduced when
>> the data had to be replicated..." The obvious but unanswered
>> question is: Why did this data have to be replicated in the first
>> place? The answer: because there was no webserver for VMS that could
>> scale to anywhere near the performance capacity that AccuWeather
>> required.
>
> I see this as a problem resulting from the absurd idea of "free software".
> Once someone gets the idea of such, then the next step is "the price is
> right, make it work", and now you're limited to whatever someone will
> choose to implement, without pay, on their own time.
>
> Needless to say, that "free application" just might not be available in
> every environment. Or when ported to another environment, might not work
> so well.
>
> I'd also wonder at just what "performance capacity" is required, or
> currently in use at for instance AccuWeather? Why would the WASD web
> server not satisfy that demand? Not saying it should, just asking where it
> might not.
>
> I don't use a web server. One of my shortcomings, perhaps, perhaps not.
> To me a web server is just a listener socket, accepting connection
> requests, with some capability to do something with some of the data sent
> by a client. Perhaps over-simplified, but, that's what it does.

The big different is the way you present data to the user. Much better
ways to do that on a web browesr then on a VT-screen...

Then you have all the FORM processing going in within the webserver
before the applications sees the call. And other stuff...

>
> With sufficient specifications, I cannot see why a web server could not be
> implemented on VMS to satisfy any requirements. Even a large cluster
> running multiple copies of the web server to meet the volume of connection
> requests. (No, I haven't volunteered to write such.)

Why write another web server to do what WASD already does (well) ?
WASD is fully cluster-ready, as far as I know.

>
> Also, you cannot solve a problem, unless you know what the problem is. I'd
> be curious as to where and why VMS based web servers do not satisfy the
> requirements?
>
> The subject interests me because a cousin of the web server, a web service,
> has greatly influenced the applications he's working with.
>
> In the old days, people would sit at terminals, with phones, and takes
> customer orders over the phone. (Dave shows his extreme age.) We have
> implemented "web services" and a protocol to enable customers to submit
> orders computer-to-computer. Again, basically a listener socket, with code
> to process specific requests from clients.

This is a simple implementation using WASD and a CGI app having WS/SOAP
interface. All networking from and to the clients goes through WASD.
I did this 5-6 years ago using WASD, gSOAP, some small C-stubs and
the usual Cobol application code to process the requests.

By using WASD, there is no TCPIP programming, all networking is
already handled by the standard WASD server. No sockets. Just
config an URL that points to the application. Standard port 80.

>
> We've been discussing how to set up things with more disaster tolerance.
> There have been few problems in the past, but, things can always be made
> better. Now, I was a bit confused with the "sudden" interest in disaster
> tolerance. The question was, "what happens if we lose a day's work
> (orders)?" I wondered what the problem was. After all, we're selling a
> few lawn mower parts, how much could be lost? A couple thousand dollars
> worth of orders?
>

What, more specificaly, could make you lose any transactions at all?
Doesn't your database has the actual data and the transaction log on
separate disks at least? Can a single disk volume failure make you
lose data? I hope not.

[Craig A. Berry]

On 3/28/15 1:52 PM, David Froble wrote:
> Jess Goodman wrote:
>
>> "...bottlenecks and performance problems that were introduced when
>> the data had to be replicated..." The obvious but unanswered
>> question is: Why did this data have to be replicated in the first
>> place? The answer: because there was no webserver for VMS that could
>> scale to anywhere near the performance capacity that AccuWeather
>> required.

> I'd also wonder at just what "performance capacity" is required, or
> currently in use at for instance AccuWeather? Why would the WASD web
> server not satisfy that demand? Not saying it should, just asking where
> it might not.

How do you know they didn't evaluate WASD and found that whatever
performance advantages it offered over Apache weren't enough compared to
what they could get with other web servers on other platforms?

> With sufficient specifications, I cannot see why a web server could not
> be implemented on VMS to satisfy any requirements. Even a large cluster
> running multiple copies of the web server to meet the volume of
> connection requests. (No, I haven't volunteered to write such.)

Consider how many VMS-related performance disadvantages have been
reported here, including network slowness, possibly at the ethernet
level as well as the TCP/IP stack, file I/O slowness, and the fact that
ten years ago Itanium was barely starting to catch up with where Alpha
had ended a few years previously, both in cpu and memory bandwidth.

Consider also that the problem may not have been between the web server
and web clients but between the web server and the database back end,
where the necessity of using socket communication instead of named pipes
or Unix domain sockets might well have given a poor showing.

Some of these things may have been mitigated in the 8.3 and 8.4 releases
of VMS and more recent hardware, but there is likely still a great deal
of work to be done to make VMS a high-performing web server platform.

> Also, you cannot solve a problem, unless you know what the problem is.
> I'd be curious as to where and why VMS based web servers do not satisfy
> the requirements?

Me too, though he did say he couldn't go into more detail in this forum.

[David Froble]

Well, we all know I don't get out much, right? :-)

Hey, I got my own airport, got my own shop to build airplanes, spring is
coming, why should I bother to get out at all?

> Various Linuxes may be free to download but there are lots of people
> being paid by their big-name employers to do Linux development. The
> proportion of input from unpaid volunteers isn't quite negligible but
> it seems it's a lot smaller than you've been thinking (how does 13%
> sound?).

Linux started as something that was free, and if anyone wants to play in
that sandbox, it's my impression that they cannot charge for the OS.
Right? So, from that perspective, it's free.

But, the discussion was about web servers, I thought, and Apache seems
to be the normal choice of many. Now, who originally wrote Apache? I'm
too lazy to look that up. But if it also was something developed for a
specific platform, and the developer(s) didn't care about other
platforms, then there is a problem needing Apache on one of those other
platforms. This is where VMS is today. On the outside looking in.

I'll admit that a "for profit" developer might also limit the platform
selections, but, a for profit company might be more interested in
expanding their opportunities for profit. Someone who wrote some
software for their own use, on their platform of choice, might be much
less interested in supporting other platforms, nor interested in making
it possible for others to support other platforms. Thus my (perhaps
flawed) insight into "free software".

> The Linux Foundation does a survey of where the Linux changes come
> from, and sometimes it gets reported. Apparently you'd be surprised by
> the numbers. Perhaps others would too. So here's a recent sample.
>
> E.g. in 2013 [1], top corporate contributor was Red Hat at 10% of
> all changes. Next big names, in order, were Intel (9%), TI (4%),
> [Linaro (4%) aren't famous], Suse (3%), IBM (3%), Samsung (2%),
> and Google (2%). There are plenty more corporate names on the list,
> not all of them widely known in the land of VMS and Basic, but in
> their fields they are big names: e.g. Oracle, Broadcom, Qualcomm,
> Cisco, etc.

Back to Linux. IBM started to develop for Linux in order to sell HW and
to retain customers. Red Hat does so in order to sell services. I'd
bet just about all corporate developers of Linux do so in order to sell
something. From that perspective, stretching a bit, one might consider
Linux other than totally free.

> In 2013 the proportion of changes submitted by sole
> developers working on their own was only 13%.
>
> One well known name not showing on that list: HP. There is a category
> for "unknown", at 3%.
>
> I realise statistics are a wonderful thing, and are more often used
> for support than for illumination. But have a think about this. Linux
> isn't $$$ free, it's paid for by the customers of the companies whose
> employees are paid to contribute.

Ayep!

> Does this change anything in your picture? Not sure. But it might.

From the perspective of Apache not seeming to be a good fit on VMS, no,
it doesn't.

I'll ask, if Apache wasn't free, do you think there might be more
competitors for that market? I do. It's the "free" that stifles
competition. Who is going to spend money to develop a product to
compete with a free product? Would you?

[David Froble]

Not always ....

> Then you have all the FORM processing going in within the webserver
> before the applications sees the call. And other stuff...
>
>>
>> With sufficient specifications, I cannot see why a web server could
>> not be
>> implemented on VMS to satisfy any requirements. Even a large cluster
>> running multiple copies of the web server to meet the volume of
>> connection
>> requests. (No, I haven't volunteered to write such.)
>
> Why write another web server to do what WASD already does (well) ?
> WASD is fully cluster-ready, as far as I know.

I think so also. What puzzles me is why more people aren't using the
product. One reason is the lack of support contracts for the product.

>> Also, you cannot solve a problem, unless you know what the problem is.
>> I'd
>> be curious as to where and why VMS based web servers do not satisfy the
>> requirements?
>>
>> The subject interests me because a cousin of the web server, a web
>> service,
>> has greatly influenced the applications he's working with.
>>
>> In the old days, people would sit at terminals, with phones, and takes
>> customer orders over the phone. (Dave shows his extreme age.) We have
>> implemented "web services" and a protocol to enable customers to submit
>> orders computer-to-computer. Again, basically a listener socket, with
>> code
>> to process specific requests from clients.
>
> This is a simple implementation using WASD and a CGI app having WS/SOAP
> interface. All networking from and to the clients goes through WASD.
> I did this 5-6 years ago using WASD, gSOAP, some small C-stubs and
> the usual Cobol application code to process the requests.
>
> By using WASD, there is no TCPIP programming, all networking is
> already handled by the standard WASD server. No sockets. Just
> config an URL that points to the application. Standard port 80.

Jan-Erik, what am I to do with you ???

Perhaps YOU don't have to do any TCP/IP and socket work, but what do you
think is at the bottom of WASD? That's right, SOCKETS! I've discussed
some things with Mark Daniel, and he gave me some hints on how to pass a
socket off to another process. Something that I've yet to do, as other
things got in the way.

As for using a bunch of "middleware", some of it I prefer to call
"bloatware", making things much more complex, and slow.

What I've done is VERY fast.

It also allows us to do exactly what we want, without possibly having to
jump through some hoops to get some middleware to fit our needs.

Perhaps the problems at AccuWeather is the use of such middleware and
being able to get the desired through-put. Don't know. Would be
interested in just what their problems are / were.

>> We've been discussing how to set up things with more disaster tolerance.
>> There have been few problems in the past, but, things can always be
>> made
>> better. Now, I was a bit confused with the "sudden" interest in disaster
>> tolerance. The question was, "what happens if we lose a day's work
>> (orders)?" I wondered what the problem was. After all, we're selling a
>> few lawn mower parts, how much could be lost? A couple thousand dollars
>> worth of orders?
>>
>
> What, more specificaly, could make you lose any transactions at all?
> Doesn't your database has the actual data and the transaction log on
> separate disks at least? Can a single disk volume failure make you
> lose data? I hope not.

Your small attention span is showing. We've had these discussions
before, and I believe you're aware that we're not using a database with
such capabilities. Perhaps this is some of your sarcasm.

To directly answer your question, right now, losing all volumes of a
shadow set (we use RAID 1) would cause us to lose data. We're looking
to make this more robust, and to implement being able to break out a
shadow set for a BACKUP snapshot of the data without having applications
down for an extended period of time.

[David Froble]

Craig A. Berry wrote:
> On 3/28/15 1:52 PM, David Froble wrote:
>> Jess Goodman wrote:
>>
>>> "...bottlenecks and performance problems that were introduced when
>>> the data had to be replicated..." The obvious but unanswered
>>> question is: Why did this data have to be replicated in the first
>>> place? The answer: because there was no webserver for VMS that could
>>> scale to anywhere near the performance capacity that AccuWeather
>>> required.
>
>> I'd also wonder at just what "performance capacity" is required, or
>> currently in use at for instance AccuWeather? Why would the WASD web
>> server not satisfy that demand? Not saying it should, just asking where
>> it might not.
>
> How do you know they didn't evaluate WASD and found that whatever
> performance advantages it offered over Apache weren't enough compared to
> what they could get with other web servers on other platforms?

I don't know that. Nor do I know if this was one of the things he could
not discuss.

>> With sufficient specifications, I cannot see why a web server could not
>> be implemented on VMS to satisfy any requirements. Even a large cluster
>> running multiple copies of the web server to meet the volume of
>> connection requests. (No, I haven't volunteered to write such.)
>
> Consider how many VMS-related performance disadvantages have been
> reported here, including network slowness, possibly at the ethernet
> level as well as the TCP/IP stack, file I/O slowness, and the fact that
> ten years ago Itanium was barely starting to catch up with where Alpha
> had ended a few years previously, both in cpu and memory bandwidth.

All true ..

> Consider also that the problem may not have been between the web server
> and web clients but between the web server and the database back end,
> where the necessity of using socket communication instead of named pipes
> or Unix domain sockets might well have given a poor showing.

I'd assume if using VMS then the database back end would also be on VMS.
I doubt they used VMS just for the web server.

> Some of these things may have been mitigated in the 8.3 and 8.4 releases
> of VMS and more recent hardware, but there is likely still a great deal
> of work to be done to make VMS a high-performing web server platform.

Agreed ..

>> Also, you cannot solve a problem, unless you know what the problem is.
>> I'd be curious as to where and why VMS based web servers do not satisfy
>> the requirements?
>
> Me too, though he did say he couldn't go into more detail in this forum.
>

But we can be curious ..

[Simon Clubley]

On 2015-03-29, David Froble <da...@tsoft-inc.com> wrote:

> johnwa...@yahoo.co.uk wrote:
>> Various Linuxes may be free to download but there are lots of people
>> being paid by their big-name employers to do Linux development. The
>> proportion of input from unpaid volunteers isn't quite negligible but
>> it seems it's a lot smaller than you've been thinking (how does 13%
>> sound?).
>
> Linux started as something that was free, and if anyone wants to play in
> that sandbox, it's my impression that they cannot charge for the OS.
> Right? So, from that perspective, it's free.
>

Wrong. Totally, totally wrong. Sorry David. :-)

You can charge as much as you like for a Linux distribution but what
you can't do is to refuse to ship the GPL source code used to build
that binary distribution to your customers.

Even though you can try to encourage them not to, you also can't
stop your customers from further redistributing the source code from
the GPL parts of your distribution.

This is how Scientific Linux and Centos can exist. They take the RHEL
source code, strip out any remaining RH copyrighted material and
build a new distribution based around the RHEL source code.

It's also why, when you buy (say) home routers with GPL based software
in them, the router comes with a little slip of paper saying how you
can obtain the source code for the GPL parts of the software in the
router.

>
> I'll ask, if Apache wasn't free, do you think there might be more
> competitors for that market? I do. It's the "free" that stifles
> competition. Who is going to spend money to develop a product to
> compete with a free product? Would you?

No. If there wasn't a free option, the existing commercial options
would cost a lot more and have a fraction of the functionality.
The free option keeps the commercial vendors on their toes and
requires them to product a product better than the free one.

If a particular product range can't be enhanced over a free version
then you can still make money by selling top quality support for
the free version.

Simon.

--
Simon Clubley, clubley@remove_me.eisner.decus.org-Earth.UFP
Microsoft: Bringing you 1980s technology to a 21st century world

[Stephen Hoffman]

On 2015-03-29 04:08:32 +0000, David Froble said:

> Linux started as something that was free, and if anyone wants to play
> in that sandbox, it's my impression that they cannot charge for the OS.
> Right? So, from that perspective, it's free.

That depends on the license involved.

The Linux kernel uses GPLv2. The Linux userland uses a mix of
licenses, usually including GPLv2 and GPLv3. GPLv2 does allow folks to
charge for the product and to charge for product support and there can
be proprietary extensions and/or logos involved in some packages,
though the vendors are required to offer the source code, if the source
code is not provided with the purchase. For a commercial example of
such commercial software, RedHat Enterprise Linux. Dual licenses are
also in use in some products, where the vendor releases some or all of
of the source code under GPLv2 or GPLv3, and also either releases or
offers to release equivalent or differing software distributions and
other packaging configurations under a different license.

Apache does not use GPL, however. Apache uses the Apache license.

A third license is the MIT/BSD-style license, and there are many other
open-source licenses and variations around.

> But, the discussion was about web servers, I thought, and Apache seems
> to be the normal choice of many.

Nginx and some other web servers are also used widely.
<http://en.wikipedia.org/wiki/Comparison_of_web_server_software>

> I'll ask, if Apache wasn't free, do you think there might be more
> competitors for that market? I do. It's the "free" that stifles
> competition. Who is going to spend money to develop a product to
> compete with a free product? Would you?

Microsoft IIS is in use at a number of sites, as are some other
commercial options.

It's also common to see competition among other "free" packages. Nginx
was launched because of issues with Apache, for instance.

In a manner of consideration, the open-source versions of the various
packages are the trial versions. There's also not much of a market for
piracy of these open-source versions, but then details such as software
piracy and ensuring distribution integrity are fodder for other
discussions.

But yes, the advent of open source certainly changes the market for web
servers, compilers, operating systems and other products, whether in
terms of the obvious pricing floor, or the expectations of
compatibility, the user interfaces, or other details. There are still
cases — such as the management front-end and user interfaces that Apple
provides for Apache and other open source, or the HP OneView tools —
where there can be profits with open-source.

[johnso...@gmail.com]

On Saturday, March 28, 2015 at 2:45:49 PM UTC-4, David Froble wrote:

> To me a web server is just a listener socket, accepting connection
> requests, with some capability to do something with some of the data
> sent by a client. Perhaps over-simplified, but, that's what it does.

That's a fair summary.

> With sufficient specifications, I cannot see why a web server could not
> be implemented on VMS to satisfy any requirements. Even a large cluster
> running multiple copies of the web server to meet the volume of
> connection requests. (No, I haven't volunteered to write such.)

In my experience, that code is always just slower and less scalable than
the same counterpart on Linux. I've participated in such an in house
project (home grown http server on VMS) and it would always have higher
latency than its counterpart on Linux.

After years of disbelief and frustration, I spent a few weeks really digging
into it, and from what I could observe, much of the blame really sat at the
hands of the TCP/IP stack. Interestingly enough, both Multinet and
TCP/IP services came up short.

I believe its possible to fix them, but it will take time.

EJ

[Stephen Hoffman]

On 2015-03-29 13:11:09 +0000, johnso...@gmail.com said:

> In my experience, that code is always just slower and less scalable
> than the same counterpart on Linux. I've participated in such an in
> house project (home grown http server on VMS) and it would always have
> higher latency than its counterpart on Linux.
>
> After years of disbelief and frustration, I spent a few weeks really
> digging into it, and from what I could observe, much of the blame
> really sat at the hands of the TCP/IP stack. Interestingly enough, both
> Multinet and TCP/IP services came up short.
>
> I believe its possible to fix them, but it will take time.

Ayup. The driver stack, too. Linux has spent a whole lot of time
optimizing how that works. Fewer and preferably no buffer copies,
lighter-weight I/O operations, etc. Beyond any kernel changes and
renovations within the TCP/IP Services stack — and I still think VSI
might well replace that with a Process IP stack, but I digress —
improving this performance might be disruptive and/or require
application changes, too.

Each generation of faster NIC means less time for the host software to
"dawdle", too. Less time for the host to even do other things.

Related: <https://lwn.net/Articles/629155/>

Web servers can involve either some rather big hosts, or can involve
pools of smaller hosts, or a combination. Having lots of idle serves
is a waste, which is one of the reasons why there are folks quite fond
of using public or private clouds here — HP Helion / OpenStack, or
otherwise — as these seek to avoid over=provisioning your servers for
expected peak loads. Think of how "Galaxy" could migrate cores among
instances, but implemented with whole fleets of servers, and not
limited to the cores within a single Alpha box.

Some other issues that arise involve managing software deployments and
large-configuration management — even in a cluster, VMS is somewhat
clunky here, and VMS just isn't very good at these sorts of tasks
beyond the scope of a cluster — and then there's scaling — somebody
will have to figure out how to deal with large clusters of lots of
boxes and potentially of lots of cores — and then there are discussions
of pricing — clustering is comparatively expensive. Then there's the
distributed database discussion, because some folks will want to be
able to support multiple clusters and/or multiple data centers and some
sort of failover.

[Craig A. Berry]

On 3/28/15 11:08 PM, David Froble wrote:

> I'll admit that a "for profit" developer might also limit the platform
> selections, but, a for profit company might be more interested in
> expanding their opportunities for profit. Someone who wrote some
> software for their own use, on their platform of choice, might be much
> less interested in supporting other platforms, nor interested in making
> it possible for others to support other platforms. Thus my (perhaps
> flawed) insight into "free software".

Maybe. Or maybe the open source developers are motivated by pride rather
than profit and their values include portability for its own sake.

[Kerry Main]

> -----Original Message-----
> From: Info-vax [mailto:info-vax...@info-vax.com] On Behalf Of
> johnso...@gmail.com
> Sent: 29-Mar-15 9:11 AM
> To: info...@info-vax.com
> Subject: Re: [New Info-vax] Accuweather new contract
>

Wrt to Multinet, I am wondering if your testing was before the most
recent version V5.4 and if it included things like fastpath in the testing:

>From Multinet V5.4 background info:

* The INET and BG drivers have been modified to use buffered I/O,
which delivers higher throughput on multiprocessor systems. (Note
that users running on OpenVMS 8.x may need to increase their non-
paged pool SYSGEN variables.)

* Support for FASTPATH to increase throughput when exchanging
packets with the Ethernet controller(s) on OpenVMS 8.x systems.

If it was Apache on OpenVMS, then I could see how another platform
might outperform Apache on OpenVMS. As I understand it, Apache
uses a model of a parent process controlling many child processes
(one per request). OpenVMS process creation is more expensive in
Terms of resources consumed. UNIX process creation is different and
does not contain the overhead of OpenVMS process creation (apples
and oranges)

WASD uses a dedicated server process - not unlike what a transaction
processing system uses i.e. dedicated processes that handle many
incoming connections in a more efficient model

>From a past WASD presentation from Mark Daniel:
"WASD uses a single process and ASTs to enable an event-driven
(mainly I/O but with some timer queue) multiple request concurrency
This model could be referred to as lightweight-threading. These are
very lightweight in the sense they are VMS' native threading model,
almost negligible servicing cost and certainly containing none of the
thread-management overhead of something like POSIX Threads or
a process context. The WASD conservative approach to resource
consumption in this respect is often a significant factor in preference
over other approaches."

[Simon Clubley]

On 2015-03-29, Simon Clubley <clubley@remove_me.eisner.decus.org-Earth.UFP> wrote:
> On 2015-03-29, David Froble <da...@tsoft-inc.com> wrote:
>> johnwa...@yahoo.co.uk wrote:
>>> Various Linuxes may be free to download but there are lots of people
>>> being paid by their big-name employers to do Linux development. The
>>> proportion of input from unpaid volunteers isn't quite negligible but
>>> it seems it's a lot smaller than you've been thinking (how does 13%
>>> sound?).
>>
>> Linux started as something that was free, and if anyone wants to play in
>> that sandbox, it's my impression that they cannot charge for the OS.
>> Right? So, from that perspective, it's free.
>>
>
> Wrong. Totally, totally wrong. Sorry David. :-)
>

There was an additional thing at the back of my mind when I made that
statement which is why I replied as firmly as I did. :-)

ACT is an Ada compiler company known in certain circles for the quality
of their work. The compiler they wrote is based on gcc and hence is
covered by the GPL.

They actively develop and enhance this compiler; they are not simply
a Red Hat style distributor. They maintain their own branch of gcc
to a safety critical standard:

http://www.adacore.com/gnatpro-safety-critical

My point is that they are producing high quality software which is
GPL based at it's core.

As per the GPL, they are only required to give the source code from
this branch to their paying customers, but they regularly push some
changes back into the public FSF branch of gcc.

BTW, if the prices I have seen mentioned for their support contracts are
accurate, then they are expensive - and people willingly pay those prices.

[Kerry Main]

> -----Original Message-----
> From: Info-vax [mailto:info-vax...@info-vax.com] On Behalf Of
> David Froble
> Sent: 29-Mar-15 12:09 AM
> To: info...@info-vax.com

> Subject: Re: [New Info-vax] OT(?): Linux: developed by corporates.
> *NOT* developed by unpaid volunteers.
>

[snip...]

> >
> > I realise statistics are a wonderful thing, and are more often used
> > for support than for illumination. But have a think about this. Linux
> > isn't $$$ free, it's paid for by the customers of the companies whose
> > employees are paid to contribute.
>
> Ayep!
>
> > Does this change anything in your picture? Not sure. But it might.
>
> From the perspective of Apache not seeming to be a good fit on VMS,
> no,
> it doesn't.
>
> I'll ask, if Apache wasn't free, do you think there might be more
> competitors for that market? I do. It's the "free" that stifles
> competition. Who is going to spend money to develop a product to
> compete with a free product? Would you?

Ask all of those backup vendors who are doing well selling backup
products for Windows and Linux when those platforms includes "free"
platform utilities to do this.

Same think could be stated for Wintel/Linux batch schedulers.

Cost is one factor, but if there is a commercial product that has much
more functionality, then most med-large Cust's will pay the extra cost.

It's the reason why there are still many Multinet/TCPware Customers
on OpenVMS when the native stack is "free".

[David Froble]

Simon Clubley wrote:
> On 2015-03-29, David Froble <da...@tsoft-inc.com> wrote:
>> johnwa...@yahoo.co.uk wrote:
>>> Various Linuxes may be free to download but there are lots of people
>>> being paid by their big-name employers to do Linux development. The
>>> proportion of input from unpaid volunteers isn't quite negligible but
>>> it seems it's a lot smaller than you've been thinking (how does 13%
>>> sound?).
>> Linux started as something that was free, and if anyone wants to play in
>> that sandbox, it's my impression that they cannot charge for the OS.
>> Right? So, from that perspective, it's free.
>>
>
> Wrong. Totally, totally wrong. Sorry David. :-)

I could be wrong. I've been wrong before. Same old territory for me.

> You can charge as much as you like for a Linux distribution but what
> you can't do is to refuse to ship the GPL source code used to build
> that binary distribution to your customers.
>
> Even though you can try to encourage them not to, you also can't
> stop your customers from further redistributing the source code from
> the GPL parts of your distribution.
>
> This is how Scientific Linux and Centos can exist. They take the RHEL
> source code, strip out any remaining RH copyrighted material and
> build a new distribution based around the RHEL source code.
>
> It's also why, when you buy (say) home routers with GPL based software
> in them, the router comes with a little slip of paper saying how you
> can obtain the source code for the GPL parts of the software in the
> router.

Can you point out in what you've written that contradicts what I wrote?
You're throwing out some information, which I have no problem with,
but I don't see how it contradicts in any way what I (obviously
erroneously) wrote.

Specifically, does, or can, anyone charge a fee for the OS ?

>> I'll ask, if Apache wasn't free, do you think there might be more
>> competitors for that market? I do. It's the "free" that stifles
>> competition. Who is going to spend money to develop a product to
>> compete with a free product? Would you?
>
> No. If there wasn't a free option, the existing commercial options
> would cost a lot more and have a fraction of the functionality.
> The free option keeps the commercial vendors on their toes and
> requires them to product a product better than the free one.

That's a good argument, but ....

Are there commercial versions of Apache? Specifically, those with a
license (or other) fee ?

> If a particular product range can't be enhanced over a free version
> then you can still make money by selling top quality support for
> the free version.

Yes, that's what some vendors are doing. But, that's charging for the
support, not for the product, right ?

We all know I don't get out much, but it's been my impression that GPL
code cannot be resold for a fee. Is this correct?

[David Froble]

If so, how much has that "pride" helped their creations run well on VMS?

[David Froble]

Kerry Main wrote:

> Ask all of those backup vendors who are doing well selling backup
> products for Windows and Linux when those platforms includes "free"
> platform utilities to do this.

Oh, I must be really dense. Can you tell me how to backup and restore a
weendoze system disk? Using the backup provided with weendoze ..

Without that, you don't really have a usable backup, do you?

[Simon Clubley]

On 2015-03-29, David Froble <da...@tsoft-inc.com> wrote:

> Kerry Main wrote:
>
>> Ask all of those backup vendors who are doing well selling backup
>> products for Windows and Linux when those platforms includes "free"
>> platform utilities to do this.
>
> Oh, I must be really dense. Can you tell me how to backup and restore a
> weendoze system disk? Using the backup provided with weendoze ..
>

One option is the exact same way you do it on VMS; you use a tool which
provides a Windows/Linux version of standalone backup.

[Simon Clubley]

On 2015-03-29, David Froble <da...@tsoft-inc.com> wrote:

> Simon Clubley wrote:
>> If a particular product range can't be enhanced over a free version
>> then you can still make money by selling top quality support for
>> the free version.
>
> Yes, that's what some vendors are doing. But, that's charging for the
> support, not for the product, right ?
>
> We all know I don't get out much, but it's been my impression that GPL
> code cannot be resold for a fee. Is this correct?

It _can_ legally be resold for a fee; it's just that most people choose
not to do that or use a dual licence approach.

In embedded devices however, the cost of integrating Linux into the
product is going to be included in the product.

Also, consider vendors like Dell who can sell their systems pre-packaged
with Linux and have done the integration work to make sure Linux works
on those systems out of the box.

[David Froble]

Simon Clubley wrote:
> On 2015-03-29, David Froble <da...@tsoft-inc.com> wrote:
>> Simon Clubley wrote:
>>> If a particular product range can't be enhanced over a free version
>>> then you can still make money by selling top quality support for
>>> the free version.
>> Yes, that's what some vendors are doing. But, that's charging for the
>> support, not for the product, right ?
>>
>> We all know I don't get out much, but it's been my impression that GPL
>> code cannot be resold for a fee. Is this correct?
>
> It _can_ legally be resold for a fee; it's just that most people choose
> not to do that or use a dual licence approach.

Hmmm ....

I'm going to question you on this statement for several reasons.

I can see charging for distribution media ...
I can see charging for additions to the GPL code ...
I can see charging for services ...
I can see charging for support ...

But I had the very strong impression that the code under the GPL license
could not be sold. Am I wrong?

> In embedded devices however, the cost of integrating Linux into the
> product is going to be included in the product.

Yes, and that is not, strictly speaking, charging for the GPL software.
Only if the price was broken out and there was a piece for the GPL
software would you say that the GPL software was "sold". Would you agree?

> Also, consider vendors like Dell who can sell their systems pre-packaged
> with Linux and have done the integration work to make sure Linux works
> on those systems out of the box.

Again, selling HW. Does Dell have a separate line item for the software?

[David Froble]

Simon Clubley wrote:
> On 2015-03-29, David Froble <da...@tsoft-inc.com> wrote:
>> Kerry Main wrote:
>>
>>> Ask all of those backup vendors who are doing well selling backup
>>> products for Windows and Linux when those platforms includes "free"
>>> platform utilities to do this.
>> Oh, I must be really dense. Can you tell me how to backup and restore a
>> weendoze system disk? Using the backup provided with weendoze ..
>>
>
> One option is the exact same way you do it on VMS; you use a tool which
> provides a Windows/Linux version of standalone backup.
>
> Simon.
>

But, that is not part of the standard weendoze distribution. It's a
third party (or even MS) add-on.

[John E. Malmberg]

On 3/29/2015 2:40 PM, David Froble wrote:
>
> But I had the very strong impression that the code under the GPL license
> could not be sold. Am I wrong?

Yes.

You can sell the code under the GPL license. You must provide to the
customer the source code either directly or available as a download site
for some period of time as specified in the license details.

You can not restrict what your customer does with the GPL code you sold
them. They can put the source code on their web site for free download,
or can sell it to others.

Or the customer could decide to not to redistribute to others the code
they paid money for.

Regards,
-John
wb8...@qsl.network
Personal Opinion Only

[Kerry Main]

> -----Original Message-----
> From: Info-vax [mailto:info-vax...@info-vax.com] On Behalf Of
> David Froble
> Sent: 29-Mar-15 12:51 PM
> To: info...@info-vax.com
> Subject: Re: [New Info-vax] OT(?): Linux: developed by corporates.
> *NOT* developed by unpaid volunteers.
>

As per Mr. Google:
http://tinyurl.com/windows-backup-1

[John E. Malmberg]

Recent Microsoft Windows, from at least Vista onwards provide a backup
tool that can create "image" backups.

The distribution DVD/Thumbdrive can be booted and restore that backup.

For most retail PCs, the restored backup can only be restored on
hardware that is running almost the exact same BIOS as what the backup
was taken on. Works for upgrading the hard drive.

For changing the hardware, then you have to install Windows and then
create all the accounts before restoring the user's files from the old
backup.

Microsoft tools for moving or "upgrading the OS" to a new PC assume that
you are always moving from a running older system to the new system.

The tools and their on-line instructions that can be easily found do not
handle the case where you have salvaged the hard drive from a dead
system and want to use it for the new system.

Regards,
-John

[JF Mezei]

re: improving the IP stack.

Some newer OS have IP stacks that are able to handle TCP much more
efficiently, not only offloading work to the ethernet card, but also
sending ACKs for multiple packets instead of ACKs for each packet.

aka: if you have received sequence number 562 and 563, you can ACK 563
and this implicitely says you have received all packets until 563, and
this includes 562. For fast data transfers, this ends up greatly
reducing ACK processing and traffic on the uplink.

[JF Mezei]

On 15-03-29 12:51, David Froble wrote:

> Oh, I must be really dense. Can you tell me how to backup and restore a
> weendoze system disk? Using the backup provided with weendoze ..

Proof that Mr VAXman is more dangerous than me. He has managed to infect
more people with his "weendoze", while I have not been able to infect
anyone with my "8086"

:-)

[johnwa...@yahoo.co.uk]

You need to stay in more and play with Windows :) Or maybe I need to get
out more - I'm working on it :)

In the early daze of Windows NT, MS bundled a cut-down version of a
commercial backup tool. I forget which one it was (this was when I
had NT at home but not at work, and I had the joys of a Travan tape,
before paying £150 or so for a DVD writer and/or inheriting a discarded
TLZ04).

This software ran under Windows, which didn't fill me with joy, for all
the same reasons that have been written wrt /IGNORE=INTERLOCK.

So my preferred option in the early daze was actually Norton Ghost
(think Standalone Backup). It used a DOS-based tool for both backup and
restore. Full disk (or partition) image only, though file by file
restore under Windows was a later development.

In recent (and indeed not so recent) years the idea of shutting down
to take a backup has fallen out of favour. Sadly, Ghost eventually fell
in with the "do it live" gang and I stopped using it.

Around the same time, "enterprise" backup packages were (re-)designed
to work with "backup agents" for major "enterprise" software packages,
which magically ensure that the backup package gets to see a consistent
set of data. Allegedly.

More recently still, MS have rediscovered 'StorageWorks Virtual Replicator
for Windows NT'. SVR was a bit like Host Based Volume Shadowing, except
copy-on-write as you go along, rather than copy a whole device, so you
could do useful things with it with a relatively small amount of 'free'
space. And MS have provided a way for applications to control it.

This seems to have resulted in there being a backup package in (some
flavours of?) Windows 7. And snapshotting is also used by non-MS packages
including the one I am using this week (Easeus, which seems much more
robust than the MS-supplied one). Package takes a snapshot (which doesn't
occupy any space unless writes occur before the backup completes) and the
backup data is copied from the point-in-time image of the disk(s).

But as with dismembering a shadow set on VMS and taking a backup from
that, the snapshot doesn't guarantee that the backup is self-consistent.
Only the applications (and OS components) know when it's *safe* to take
the snapshit.

And there's more. I've used the MS-supplied backup on a number of Win7
systems. Eventually, it breaks, and refuses to take a backup. Ever again.

There never seems to be a definitive answer as to why, and obviously
the acceptable answer nowadays is "don't diagnose, don't fix, just
reinstall the OS and hope it doesn't happen again". At least it's quick.

My best guess for the failures I've seen is that they result from the
pre-reserved "copy-on-write" space becoming full, and either there's
no mechanism to extend it, or (more likely) sometimes it doesn't work.
Other suggestions very welcome (but not necessarily here).

Note: the above mostly talks about taking a backup.

Restore is a different kettle of worms. I'll stick with the low end
packages I know, not the "enterprise" stuff (some of which are, at least
in the UK and Europe, supported by HP).

Selective file by file restore with some of these things is easy - an
explorer-like interface is typically available, but you may have to work
out where the file you are looking for is available. (some flavours of?)
Windows 7 have a built in "previous versions" feature which work with the
built-in Win7 backup. If it was trustworthy, it might simplify some of
this. See above - not trustworthy imo, so I don't use this feature.

"Bare metal" restore typically involves a bootable recovery disk with
an application that knows how to read the "save set" and restore it. Some
packages even claim to be able to restore a Windows setup onto different
hardware than the original system.

Products I've tried post-Ghost (which was DOS based) have typically used
a Linux-based recovery setup, as Windows isn't the right tool for that
job (one size does not fit all, remember - unless you're Microsoft or
Microsoft-dependent).

Any clearer?


TL;DR: yes MS have shipped a backup package with Windows at various
times including NT3, NT4, and Win7 (also maybe Win2K, maybe WinXP). Have
they been useful? Somewhat. Is the Win7 package useful? Probably not
very, in my experience.

[johnso...@gmail.com]

On Sunday, March 29, 2015 at 11:10:05 AM UTC-4, Kerry Main wrote:

> Wrt to Multinet, I am wondering if your testing was before the most
> recent version V5.4 and if it included things like fastpath in the testing:

This would have been 2010 so probably not. I can't remember if fastpath
was an option that was available or tested.

The comparison strategy is actually very simple. Instead of focusing on TCP
and all of its well intended options, I focused on UDP since that enabled a
simpler baseline.

My basic mode of comparison was... how long does it take to send a single
80 byte UDP packet. I measured that by doing that in a tight loop. It was pretty
easy to write a portable version that ran on both Linux and VMS. For grins
I also write a version that used the QIO$ interface. That improved things a little
bit, but Linux still won by a wide margin.

EJ

[Kerry Main]

> -----Original Message-----
> From: Info-vax [mailto:info-vax...@info-vax.com] On Behalf Of
> johnso...@gmail.com
> Sent: 29-Mar-15 4:49 PM
> To: info...@info-vax.com
> Subject: Re: [New Info-vax] Accuweather new contract
>

Not saying that Linux might be marginally faster, but usually when there Is
a "wide margin", there is usually something else under pinning the results.

When network results like this are seen, I usually think of the mis-matched
full duplex auto-config issue with older network gear, NIC's and drivers. This
was quite common a few years back. A number of driver changes were made
to try and minimize these challenges, but the end result was a half-duplex
config that saw OpenVMS network performance literally cut in half.

The easy fix was to hard code the speeds via OS SW or the console to match
the network gear.

I still hear of these issues happening at some sites today e.g. "why does my
OpenVMS FTP xfer take so much longer than may XYZ platform?" This is not
the only possible reason, but is usually one of the first things I ask the local
resource to check.

[Kerry Main]

> -----Original Message-----
> From: Info-vax [mailto:info-vax...@info-vax.com] On Behalf Of
> Kerry Main
> Sent: 29-Mar-15 11:08 AM
> To: comp.os.vms to email gateway
> Subject: Re: [New Info-vax] Accuweather new contract
>

> > -----Original Message-----
> > From: Info-vax [mailto:info-vax...@info-vax.com] On Behalf Of
> > johnso...@gmail.com
> > Sent: 29-Mar-15 9:11 AM
> > To: info...@info-vax.com
> > Subject: Re: [New Info-vax] Accuweather new contract
> >

> Wrt to Multinet, I am wondering if your testing was before the most
> recent version V5.4 and if it included things like fastpath in the testing:
>

More background on WASD web serving from the HP site:

http://h71000.www7.hp.com/openvms/journal/v7/wasd.html

Also talks about the issues associated with Web Apps using a process
per connection model on OpenVMS (Apache)

In a nutshell - instead of trying to trying to solve a problem by copying
a UNIX model on OpenVMS (e.g. Apache), WASD uses OpenVMS
mechanisms to solve the problem more efficiently.

[Craig A. Berry]

On 3/29/15 6:03 PM, Kerry Main wrote:
>> -----Original Message-----
>> From: Info-vax [mailto:info-vax...@info-vax.com] On Behalf Of
>> johnso...@gmail.com
>> Sent: 29-Mar-15 4:49 PM
>> To: info...@info-vax.com
>> Subject: Re: [New Info-vax] Accuweather new contract

>> My basic mode of comparison was... how long does it take to send a
>> single
>> 80 byte UDP packet. I measured that by doing that in a tight loop. It was
>> pretty
>> easy to write a portable version that ran on both Linux and VMS. For
>> grins
>> I also write a version that used the QIO$ interface. That improved things
>> a little
>> bit, but Linux still won by a wide margin.
>>
>> EJ
>
> Not saying that Linux might be marginally faster, but usually when there Is
> a "wide margin", there is usually something else under pinning the results.
>
> When network results like this are seen, I usually think of the mis-matched
> full duplex auto-config issue with older network gear, NIC's and drivers.

Kerry, please read more details of Eric's careful analysis in the
archives of this group before posting such nonsense.

[Simon Clubley]

On 2015-03-29, JF Mezei <jfmezei...@vaxination.ca> wrote:
> re: improving the IP stack.
>
> Some newer OS have IP stacks that are able to handle TCP much more
> efficiently, not only offloading work to the ethernet card, but also
> sending ACKs for multiple packets instead of ACKs for each packet.
>

Hmm, this has been a standard part of TCP right from the beginning
(RFC 793).

> aka: if you have received sequence number 562 and 563, you can ACK 563
> and this implicitely says you have received all packets until 563, and
> this includes 562. For fast data transfers, this ends up greatly
> reducing ACK processing and traffic on the uplink.
>

Are you saying there are stacks around today which don't actually do
this ?

[Craig A. Berry]

On 3/29/15 10:07 AM, Kerry Main wrote:
>> -----Original Message-----

> If it was Apache on OpenVMS, then I could see how another platform
> might outperform Apache on OpenVMS. As I understand it, Apache
> uses a model of a parent process controlling many child processes
> (one per request).

As far as I know, that hasn't been the case since Apache 1.x. I believe
the most common model now is to run it as a so-called pre-forking server
where multiple worker processes are created at start-up time and are all
ready to handle connections. There is a new thread (but not a new
process) started for each connection.

> WASD uses a dedicated server process - not unlike what a transaction
> processing system uses i.e. dedicated processes that handle many
> incoming connections in a more efficient model
>
>>From a past WASD presentation from Mark Daniel:
> "WASD uses a single process and ASTs to enable an event-driven
> (mainly I/O but with some timer queue) multiple request concurrency
> This model could be referred to as lightweight-threading. These are
> very lightweight in the sense they are VMS' native threading model,
> almost negligible servicing cost and certainly containing none of the
> thread-management overhead of something like POSIX Threads or
> a process context. The WASD conservative approach to resource
> consumption in this respect is often a significant factor in preference
> over other approaches."

This is a great way to make efficient use of a single processor. It
appears that WASD also provides the ability to have multiple detached
processes coordinated via the DLM -- without that, the AST-driven model
wouldn't scale beyond one processor.

There are more details on how WASD actually works at performance
comparisons with Apache at:

<http://wasd.vsm.com.au/other/D215_WASD_Apache.ppt>

[Simon Clubley]

On 2015-03-29, David Froble <da...@tsoft-inc.com> wrote:
> Simon Clubley wrote:
>> In embedded devices however, the cost of integrating Linux into the
>> product is going to be included in the product.
>
> Yes, and that is not, strictly speaking, charging for the GPL software.
> Only if the price was broken out and there was a piece for the GPL
> software would you say that the GPL software was "sold". Would you agree?
>

New PCs in a retail store don't list a price for the Windows OS but
only the total price for the whole hardware/software package.

That doesn't change the fact you are paying for Windows on the new PC.

[Craig A. Berry]

Often better than the profit motive, but not always and it certainly
depends on the individuals involved.

But looking down your nose at open source ports really makes no sense.
If you think it does, go ahead and unplug all your VMS systems from NTP
and Kerberos and SSL and Perl and Python and Lua and Info-Zip and
BACKUP/DATA_FORMAT=COMPRESS and many others.

[Kerry Main]

As I recall, Eric's previous analysis showed OpenVMS network numbers
to be approx. half those of Linux which would be exactly what would be
he case if the autoconfig issue were present when he did the testing.

Either that or one must believe that Linux network numbers are twice
those of OpenVMS using similar network configs - something I find a bit
of a stretch.

The autoconfig issue is well known and has in various environments
impacted VAX, Alpha & Integrity. I have seen it first hand in a number
of different sites.

http://labs.hoffmanlabs.com/node/907

[JF Mezei]

On 15-03-29 23:25, Kerry Main wrote:
>
> The autoconfig issue is well known and has in various environments
> impacted VAX, Alpha & Integrity. I have seen it first hand in a number
> of different sites.

My alpha was NOT on autoconfg. port forced to 100mbps full both on alpha
and on switch.

My mac is set to have autoconfig on port.

Mac was able to significantly outperform the Alpha for FTP transfer over
the same DSL line.

I do not know when the last time engineering actually worked on the TCP
login in the IP stack (windowing acks etc). And I do not know if double
buffering really makes such a huge difference.

Since my DSL speeds are slower than the original 10mbps ethernet, you'd
think even a DSL10L would have enough CPU "umph!" to handle double
buffering fast enough that it wouldn't slow the link down. (at 100mbps
or 1gbps it probably makes a difference).

[David Froble]

John E. Malmberg wrote:
> On 3/29/2015 2:40 PM, David Froble wrote:
>>
>> But I had the very strong impression that the code under the GPL license
>> could not be sold. Am I wrong?
>
> Yes.
>
> You can sell the code under the GPL license. You must provide to the
> customer the source code either directly or available as a download site
> for some period of time as specified in the license details.

Does this apply to code you've written, or to all code under the GPL
license? I have a hard time understanding being able to take what
others have provided as free, and charging customers for that code.

[David Froble]

You make a convincing argument ..

:-)

[David Froble]

Infect ????????????????

Isn't "weendoze" the actual product name ?

[johnwa...@yahoo.co.uk]

TCP (specifically, ie not UDP) has a number of performance issues
which can show up if two valid but not entirely compatible
implementations talk to each other.

TCP over a less-than-perfect medium such as a DSL line can show this
up particularly badly in the presence of line errors leading to packet
loss.

This is not an issue of ultimate performance of the underlying stack.

It is an issue which used to be (relatively) widely known back in the
early days of mass market DSL, and used to have a decent writeup on
DSLreports (and very probably elsewhere). I can't find that writeup
right now. Basically, Linux webservers and NT clients didn't play nice
together as a result. Presumably this is long since fixed in NT, but
may not have been fixed in other OSes.

The magic keyword is "TCP slow start".

[David Froble]

johnwa...@yahoo.co.uk wrote:

Too long, but I read it anyway ....

When I can do what BACKUP /IMAGE does on VMS, with a product bundled
with the weendoze product, then I'll consider that Microsoft finally
"gets it". Until then, I'll continue with the belief that it's at best
a toy OS.

VMS can be booted and run, with some restrictions, from a CD / DVD ...

I've got a copy of Ubuntu on a DVD that is bootable and runs, with some
restrictions.

So, it can't be all that hard to be able to boot from optical media, or
USB, with enough capabilities to run a decent backup product. That
would suppose there also being a decent backup product.

[Phillip Helbig (undress to reply)]

In article <mfa9ic$ale$1...@dont-email.me>, "Craig A. Berry"

<craig...@nospam.mac.com> writes:

> > WASD uses a dedicated server process - not unlike what a transaction
> > processing system uses i.e. dedicated processes that handle many
> > incoming connections in a more efficient model

> This is a great way to make efficient use of a single processor. It
> appears that WASD also provides the ability to have multiple detached
> processes coordinated via the DLM -- without that, the AST-driven model
> wouldn't scale beyond one processor.

Don't forget the OSU server, which is also written from the ground up
with VMS in mind.

[David Froble]

Yes, there can be network issues. But perhaps a few people actually
know what they are doing, and have gotten past that.

More than a few people, knowledgeable people, have found the network
performance of VMS to be less than some competition.

To come out with such apologist statements when presented with some
facts is less than helpful. As Craig noted, you're posting useless
misdirection.

Perhaps actually addressing the problem might be more helpful ?

[Jan-Erik Soderholm]

An missconfigured half/full duplex will usualy give something
of 1:100 to 1:1000 diference. More or less a full halt.

The network is more or less useless apart from small packages.

Jan-Erik.

[John E. Malmberg]

On 3/30/2015 2:34 AM, David Froble wrote:
> John E. Malmberg wrote:
>> On 3/29/2015 2:40 PM, David Froble wrote:
>>>
>>> But I had the very strong impression that the code under the GPL license
>>> could not be sold. Am I wrong?
>>
>> Yes.
>>
>> You can sell the code under the GPL license. You must provide to the
>> customer the source code either directly or available as a download
>> site for some period of time as specified in the license details.
>
> Does this apply to code you've written, or to all code under the GPL
> license?

All code under the GPL. Note that many Linux Distributions contain some
non-GPL code in them, so you have to separate it out.

> I have a hard time understanding being able to take what
> others have provided as free, and charging customers for that code.

It is allowed. Instead of posting ported GPL code on sourceforge, I
could make it a pay for download somewhere else. I could not stop
anyone else from making it a lower cost or free download to compete with
me though.

Think of it this way. If I were the only source of compiled a GPL
package built for VMS that was needed for a niche, but profitable
market, then I could charge for it. My customers, who after paying me
my fee, could make my port free, but why should they? That could be
helping a competitor, and it could cause me to stop supporting a product
that they need.

It could be way more expensive for them to hire their own person to port
that package.


Most of the time the sellers of GPL code are selling either additional
non-GPL content, advanced access to tested GPL bug fixes, or other
support services.

But yes, some are just selling for profit what you can download for free.

Regards,
-John

[Bill Gunshannon]

In article <55185cb3$0$1575$c3e8da3$e408...@news.astraweb.com>,

JF Mezei <jfmezei...@vaxination.ca> writes:
> re: improving the IP stack.
>
> Some newer OS have IP stacks that are able to handle TCP much more
> efficiently, not only offloading work to the ethernet card, but also
> sending ACKs for multiple packets instead of ACKs for each packet.

That was tried 30 years ago. Believe it or not there was usually no
improvement in performance and in some cases it weas actually worse.
(As i have stated here before, I still have a couple of those cards
for PC's. They were almost always used in "dumb" mode with the CPU
removed. I also remember doing cards, EXCELAN I think, for the VAX.)

>
> aka: if you have received sequence number 562 and 563, you can ACK 563
> and this implicitely says you have received all packets until 563, and
> this includes 562. For fast data transfers, this ends up greatly
> reducing ACK processing and traffic on the uplink.

That has always been the case with the TCP/IP spec. I can't imagine
why any implementation would not have been taking advantage of this
unless it was just plain broken. (Unless it did not allow any kind
of windowing, but then, considering that things like Kermit were
doing this decades ago I guess that would just be another sign of how
broken an implementation was.)



bill

--
Bill Gunshannon | de-moc-ra-cy (di mok' ra see) n. Three wolves
bill...@cs.scranton.edu | and a sheep voting on what's for dinner.
University of Scranton |
Scranton, Pennsylvania | #include <std.disclaimer.h>