improving EDT

[Phillip Helbig (undress to reply)]

EDT has the nice EDIT/RECOVER feature. This is good when there is some
unexpected problem.

I have also defined a key to write a new version of the current file to
disk, so that I don't have to leave the editor. This is good when
writing source code, saving it, compiling the new version, without
leaving the editor. It also protects against making some major change
by mistake; probably, a previous version can be retrieved and used as
the basis for a new version without the mistake.

What about the two together? If I use the last feature, and there is an
unexpected crash, the journal file can't be applied to the highest
version of the edited file. If I apply it to the version of the file it
was created with, should this work in all cases, i.e. EDIT/RECOVER in
this case would give the same result as the current state of the highest
version of the file? (Obviously, if I make a big mistake and go back to
a previous version for a fresh start, I will leave the editor and the
journal file will be deleted.)

[Richard Maher]

For fuck sake how hard would it be to port Notepad++ to VMS?

I thought we also went for the wrong horse in dropping Eclipse for NetBeans?

It's just so hard to stay enthused with the level of participation here :-(

[Henry Crun]

From my experience, everything (and lots more) that can be done with EDT can be
done in TPU.
E.g. I use TPU for text eding in 8-bit Hebrew (entering text right-to-left, the
Learn function has no convenient equivalent in EDT, Spawn, etc. etc.

To use in an Edit/Compile/Test cycle you could try LSE (also based on EVE) which
has many language-specific constructs (though IIRC it costs...)

--
Mike R.
Home: http://alpha.mike-r.com/
QOTD: http://alpha.mike-r.com/qotd.php
No Micro$oft products were used in the URLs above, or in preparing this message.
Recommended reading: http://www.catb.org/~esr/faqs/smart-questions.html#before
and: http://alpha.mike-r.com/jargon/T/top-post.html
Missile address: N31.7624/E34.9691

[Phillip Helbig (undress to reply)]

In article <mt2ufd-...@Ubuntu.mike-r.com>, Henry Crun

<mi...@rechtman.com> writes:

> From my experience, everything (and lots more) that can be done with
> EDT can be done in TPU.

Perhaps. Any Turing machine can emulate another, but of course that
doesn't mean that all are equivalent. I don't like TPU reading in a
huge file if I only want to edit something at the beginning. I don't
like the cursor movement (even with EDT keypad), it is noticeably slower
than EDT (yes, I can type fast enough that TPU slows me down).

> Learn function has no convenient equivalent in EDT, Spawn, etc. etc.

LEARN is nice, and I occasionally use it. It is possible to spawn out
of EDT.

[Bob Koehler]

In article <o0hkks$e45$1...@gioia.aioe.org>, Richard Maher <maher_rj...@hotmail.com> writes:
>
> For fuck sake how hard would it be to port Notepad++ to VMS?
>

Why would anyone want to work on porting that worthless P.O.S.
to any platform?

[VAXman-]

I believe Richard was being facetious.

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

I speak to machines with the voice of humanity.

[Scott Dorsey]

<VAXman- @SendSpamHere.ORG> wrote:
>In article <F+hM4z...@eisner.encompasserve.org>, koe...@eisner.nospam.decuserve.org (Bob Koehler) writes:
>>In article <o0hkks$e45$1...@gioia.aioe.org>, Richard Maher <maher_rj...@hotmail.com> writes:
>>>
>>> For fuck sake how hard would it be to port Notepad++ to VMS?
>>>
>>
>> Why would anyone want to work on porting that worthless P.O.S.
>> to any platform?
>
>I believe Richard was being facetious.

It would be nice to ship a quick and dirty simple editor with VMS, and it
likely would be an afternoon's work to write something for novice users that
would be a lot better than Notepad.
--scott
--
"C'est un Nagra. C'est suisse, et tres, tres precis."

[Jan-Erik Soderholm]

Notepad++ != Notepad.

https://notepad-plus-plus.org/

[Bill Gunshannon]

Pico. VIM. (I think that's already available) MicroEmacs.

Guess it depends on what you like in an editor. I've used
these three and they are fine for simple editing tasks. I
think MicroEmacs had a VMS version, but that would have been
ages ago.

bill

[VAXman-]

It's already there... EDT.

[Scott Dorsey]

Bill Gunshannon <bill.gu...@gmail.com> wrote:
>On 11/16/16 10:38 AM, Scott Dorsey wrote:
>>
>> It would be nice to ship a quick and dirty simple editor with VMS, and it
>> likely would be an afternoon's work to write something for novice users that
>> would be a lot better than Notepad.
>

>Pico. VIM. (I think that's already available) MicroEmacs.

Pico would be a great start, but you need to be able to edit something other
than Stream_LF files.

>Guess it depends on what you like in an editor. I've used
>these three and they are fine for simple editing tasks. I
>think MicroEmacs had a VMS version, but that would have been
>ages ago.

They did, and I still have several users using uEmacs. It's excellent.

[hb]

On 11/16/2016 06:53 PM, Scott Dorsey wrote:
>> >Guess it depends on what you like in an editor. I've used
>> >these three and they are fine for simple editing tasks. I
>> >think MicroEmacs had a VMS version, but that would have been
>> >ages ago.
> They did, and I still have several users using uEmacs. It's excellent.
> --scott

You can find VAX and Alpha images (ue400vms-vax.zip, ue400vms-alp.zip)
as well as sources (ue400dev-vms.zip) in the decuslib, vmslt01b. There
is an announcement in vmslt01b/vu/vms-uemacs-ann.txt. It's 15 years old.
The sources have a copy right for 4.00 from 1995.

You can take whatever Linus was/is using, called uEmacs/PK. You'll find
it with the kernel sources. It is based on microemacs 3.9e, it's current
version is 4.0.15. Last change/update about two years ago. Surprisingly
it still contains VMS code, so you can build it on current versions of
VMS, but it lacks a build procedure.

[David Froble]

+10000000

[David Froble]

Ah ... yes, .... and it's called EDT ...

[Bob Koehler]

In article <o0hui1$q1r$1...@panix2.panix.com>, klu...@panix.com (Scott Dorsey) writes:
>

> It would be nice to ship a quick and dirty simple editor with VMS, and it
> likely would be an afternoon's work to write something for novice users that
> would be a lot better than Notepad.

Done. It's called TECO.

Or, by "simple" did you mean for the user?

[Jan-Erik Soderholm]

OK, I must ask...

From what I have heard from various people, Notepad++ is
a rather nice and lightweight programming editor.

Notepad, on the other hand, isn't worthless as-is, but it isn't
really a *programming* editor either.

I got the impression that some here are mistaken Notepad++
for the builtin Notepad in Windows...

Now, the major problem with Notepad++ is that it is a graphical
GUI and that doesn't play well together with VMS.

Any new edititing IDE-style environment for VMS development will
certenly not be running natively on VMS. I have used UltraEdit32
using the built-in FTP support for VMS and it works quite nice.
For better integration with such things as VMS debugger you will
probably need some more advanced c/s setup with special server
processes on VMS.

[Arne Vajhøj]

On 11/16/2016 7:49 AM, Richard Maher wrote:
> For fuck sake how hard would it be to port Notepad++ to VMS?

Very. It is supposedly build around Win32 API.

But I would expect jEdit to run on VMS. It did years ago. And
still claims that it does.

Arne

[Arne Vajhøj]

On 11/16/2016 5:15 PM, Jan-Erik Soderholm wrote:
> Den 2016-11-16 kl. 19:45, skrev David Froble:
>> Bob Koehler wrote:
>>> In article <o0hkks$e45$1...@gioia.aioe.org>, Richard Maher
>>> <maher_rj...@hotmail.com> writes:
>>>> For fuck sake how hard would it be to port Notepad++ to VMS?
>>>>
>>>
>>> Why would anyone want to work on porting that worthless P.O.S.
>>> to any platform?
>>>
>>
>> +10000000

> From what I have heard from various people, Notepad++ is
> a rather nice and lightweight programming editor.

Yes.

Notepad++ and Atom seems to be preferred general editors
these days.

Arne

[Simon Clubley]

Just had a quick look.

Notepad++ is indeed Windows only as you suggested in your other
post and Atom appears to be GUI only; there is no apparent character
cell support so you can forget about running multiple instances in
multiple xterm/PuTTY windows or over a low bandwidth SSH connection
where graphics rendering would take too long.

Simon.

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

[Arne Vajhøj]

On 11/16/2016 8:26 PM, Simon Clubley wrote:
> On 2016-11-17, Arne Vajhøj <ar...@vajhoej.dk> wrote:
>> On 11/16/2016 5:15 PM, Jan-Erik Soderholm wrote:
>>> From what I have heard from various people, Notepad++ is
>>> a rather nice and lightweight programming editor.
>>
>> Yes.
>>
>> Notepad++ and Atom seems to be preferred general editors
>> these days.
>
> Just had a quick look.
>
> Notepad++ is indeed Windows only as you suggested in your other
> post and Atom appears to be GUI only; there is no apparent character
> cell support so you can forget about running multiple instances in
> multiple xterm/PuTTY windows or over a low bandwidth SSH connection
> where graphics rendering would take too long.

Text based editors have not been in fashion for many years.

But usually those are only needed for minor work, so no
need for a very powerful editor.

I don't think there has been any new text based editor the
last 20 years or so.

Arne

[Phillip Helbig (undress to reply)]

In article <o0hui1$q1r$1...@panix2.panix.com>, klu...@panix.com (Scott
Dorsey) writes:

> It would be nice to ship a quick and dirty simple editor with VMS,

TECO is quick, dirty...but not simple.

"Quick, dirty, simple...choose any two!"

[Richard Maher]

They just don't get it Jan :-(

Just serve up a VMS file share like any other Windows/*NIX
folder/directory and it's done

[David Froble]

Phillip Helbig (undress to reply) wrote:

How about quick and simple?

[Bob Koehler]

In article <o0iuam$kpd$2...@gioia.aioe.org>, =?UTF-8?Q?Arne_Vajh=c3=b8j?= <ar...@vajhoej.dk> writes:
>
> Notepad++ and Atom seems to be preferred general editors
> these days.

Preferred by who? Nobody around here.

[Bob Koehler]

In article <o0j26t$pmo$1...@gioia.aioe.org>, =?UTF-8?Q?Arne_Vajh=c3=b8j?= <ar...@vajhoej.dk> writes:
>
> I don't think there has been any new text based editor the
> last 20 years or so.

Gee, every IDE I've seen has some text based editor built in. I
don't think I've seen the same one twice. And there seems to always
be new IDEs popping up.

Fortunately, most of them have some support for external editors.
Unfortunately, that support usually is poor.

[Jan-Erik Soderholm]

Yes, and that is part of the general problem.
Many here still lives in the past. Doesn't work
as well today as it did in the past...

[Jan-Erik Soderholm]

Of course. Not the first time either. :-)

> Just serve up a VMS file share like any other Windows/*NIX folder/directory
> and it's done
>

Yes, I did that using UltraEdit32. Works just fine. VMS files shown in the
file/directory view just as any local files. I also helped "IDM Computer
Solutions, Inc" (who produces UE32) with some minor changes to their
VMS FTP support.

And you could use Samba (or similar). Or WebDAV...

But any new/modern editing environment will quite certenly *not*
run natively on VMS...




>

[Johnny Billquist]

On 2016-11-17 08:53, Phillip Helbig (undress to reply) wrote:
> In article <o0hui1$q1r$1...@panix2.panix.com>, klu...@panix.com (Scott
> Dorsey) writes:
>
>> It would be nice to ship a quick and dirty simple editor with VMS,
>
> TECO is quick, dirty...but not simple.

I'm not sure I agree. TECO is very simple. It's just also very
unforgiving and cryptic. But it's simple.

Johnny

[David Froble]

Don't know about that. I'm thinking that everything I do today is based on the
past. As far as I can tell, things that have been around for a while still do
everything they did when first developed.

I have no problem adopting better methods, things, and such. It's what I've
always done in the past, "the past" being everything up to each letter I type.
But I will not adopt anything, just because it's new. It must show itself to be
better, better enough for me to abandon what I already have learned.

[Hans Vlems]

I'll second that.

[IanD]

Notepad++ comes with a huge number of programming language support

It is one of the most commonly suggested editors in a number of the programming forums when people start out learning programming and ask what editor should they use

One of the key strategic aspects of reviving OpenVMS was how to attract new talent, you will not do that asking people to learn any of the editors natively supplied with OpenVMS simply because new kids want portable skills, that means going with majority products and with ones they will be familiar with. After this, the framework can be embraced

I once suggested enhancements to TPU to fix a number of the little things, like box editing highlighting and undo and was shot down for the suggestion of spending effort on niech products in the grand scheme of things, on reflection of that feedback, I concluded they were right, it's time to drag OpenVMS forward and embrace new stuff, stuff that has only got better over time while OpenVMS froze itself in time

Add DCL support to notepad++ (I think there was already a module?) and invest the time going forwards

Those not wanting to embrace it and other more widely used editors and grandfather themselves, after all, they should be the ones determining if a new offering increases their productivity or not.

I know notepad++ has risen in popularity for a reason, it didn't get there through whim but through constantly meeting what people requested and demanded from a programming editor, that's free

[Kerry Main]

+1

Part of the challenge is that some like to promote todays trends
and technologies as being "current" or "next generation" or
similar hype, and anyone who talks about the past is a dinosaur.
The reality is that many of these current trends are
vendor/media/analyst hype of simple enhancements of past products
and technologies.

Most here have likely read this dinosaur article, but it bears
repeating:
http://thedailywtf.com/articles/Jurassic-Programmers-

A few examples of today's hype:
- Public Clouds: aka selective IT Outsourcing
- Private Clouds: aka internal shared services, aka IT Utility,
aka Real-Time Enterprise (Gartner), aka Adaptive Enterprise (Meta
Group, HP)
- SOA: aka DCE (distributed computing environment), aka NAS (DEC
network application services). Great in concept, and concepts do
have a place, but while getting 5 dev groups to change the way
they develop apps might be possible, good luck getting any more
than that to change and agree on who, how, what, when and why
specific services should be built, deployed and supported.
- Virtualization: while usually associated with today's server
technologies like VMware, Xen, Hyper-V, server virtualization was
in place with nPars, lPars, mainframes, application stacking for
decades before the current X86-64 focused products came on the
scene.
- Cloud Bursting: gobbligoop term created by cloud washing
promoters to mean being able to dynamically expand a computing
presence from one outsourcer (public cloud vendor) to another
outsourcer. Great in theory, but has little reality when one
considers security, data backup, data fail-back, WAN MPLS
connections (offers availability, but not latency guarantees) and
a host of other issues.
- SDN: software defined networks: great concept except no
Customer is going to throw away their existing multi-vendor,
multi-platform network infrastructure for a single vendor network
infrastructure. What about network security, support complexity
etc?

Manufacturing Customers are often viewed as Dinosaurs because
they hang on to older technologies for longer than most. In
reality, they simply understand that just because something is
shinier and sounds cool, it does not necessarily mean they should
change simply for the sake of change.

Learning from the past to plan for the future ...


Regards,

Kerry Main
Kerry dot main at starkgaming dot com

[hb]

On 11/17/2016 09:15 PM, IanD wrote:

> I once suggested enhancements to TPU to fix a number of the little
> things, like box editing highlighting and undo and was shot down for
> the suggestion of spending effort on niech products in the grand
> scheme of things, on reflection of that feedback, I concluded they
> were right, it's time to drag OpenVMS forward and embrace new stuff,
> stuff that has only got better over time while OpenVMS froze itself
> in time

Maybe not for the new kids, as it is text based: Have you looked at JED?
It runs on VMS, it has box editing, highlighting and undo. There is an
EDT keypad mode, there is a DCL, Linker options file, MMS/MMK and a VMS
system parameter mode as well as syntax highlighting for Bliss and Macro32.

For the new kids preferring Eclipse, there is a plugin for
DCL/MMS/Linker options file as well.

[Jan-Erik Soderholm]

A lot of fun stuff there...

But, we are only talkning about giving (new) programmers
on VMS a decent environment to work in. I do not understand
why you drag clouds and all that stuff into the picture.

I do not understand your post. Are you saying that editing code in
EDT/TPU is "better" then editing code in one of todays popular IDE's?

[Arne Vajhøj]

jEdit FTP and VMS also seems to work OK.

Arne

[Arne Vajhøj]

On 11/17/2016 9:08 AM, Bob Koehler wrote:
> In article <o0j26t$pmo$1...@gioia.aioe.org>, =?UTF-8?Q?Arne_Vajh=c3=b8j?= <ar...@vajhoej.dk> writes:
>>
>> I don't think there has been any new text based editor the
>> last 20 years or so.
>
> Gee, every IDE I've seen has some text based editor built in. I
> don't think I've seen the same one twice. And there seems to always
> be new IDEs popping up.

text editor => editor to edit text

text based editor => editor that is non graphical (~= runs on a 24x80 VT
terminal)

Arne

[Arne Vajhøj]

Maybe not.

But by the large developer community that exist outside
of VMS.

Arne

[Arne Vajhøj]

On 11/17/2016 3:58 PM, hb wrote:
> On 11/17/2016 09:15 PM, IanD wrote:
>> I once suggested enhancements to TPU to fix a number of the little
>> things, like box editing highlighting and undo and was shot down for
>> the suggestion of spending effort on niech products in the grand
>> scheme of things, on reflection of that feedback, I concluded they
>> were right, it's time to drag OpenVMS forward and embrace new stuff,
>> stuff that has only got better over time while OpenVMS froze itself
>> in time
>
> Maybe not for the new kids, as it is text based: Have you looked at JED?
> It runs on VMS, it has box editing, highlighting and undo. There is an
> EDT keypad mode, there is a DCL, Linker options file, MMS/MMK and a VMS
> system parameter mode as well as syntax highlighting for Bliss and Macro32.

That sounds pretty good.

> For the new kids preferring Eclipse, there is a plugin for
> DCL/MMS/Linker options file as well.

Got a link?

Arne

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

Better for who, in order to achieve what, exactly?

Someone very familiar with VMS can be very productive on
VMS, as is. What's the point of forcing folk like that to
work unproductively in an unfamiliar IDE (especially in
cases where VMS is still seen as a limited lifetime system)?

On the other hand there are lots of people out there who
are familiar with stuff like Notepad++ or Eclipse or [etc]
and it might not be a bad idea for someone (doesn't have
to be VSI) to put some thought into making these folks life
more comfortable and productive when working in and with
the VSIVMS environment.

Some of this stuff exists already and has been mentioned
here. Some of it, like the DECset client for Windows, is
as old as the hills, and thus would likely be entirely
unfashionable even if it was any good. Other products
along those lines may be available, commercially and/or
free.

More and better such stuff (or at least more visibility for
decent stuff that exists) would probably be welcome.

[David Froble]

Jan-Erik Soderholm wrote:

> But, we are only talkning about giving (new) programmers
> on VMS a decent environment to work in. I do not understand
> why you drag clouds and all that stuff into the picture.
>
> I do not understand your post. Are you saying that editing code in
> EDT/TPU is "better" then editing code in one of todays popular IDE's?

Ok, let me try explaining this way.

Does any of these "new" "wizz-bang" methods have enough advantage for me to
forget using EDT, which I pretty much got down to "muscle memory", and spend
effort (don't know how much) learning something new?

Key words are "enough advantage" ....

[Kerry Main]

> -----Original Message-----
> From: Info-vax [mailto:info-vax...@rbnsn.com] On Behalf
> Of Jan-Erik Soderholm via Info-vax
> Sent: 17-Nov-16 4:28 PM
> To: info...@rbnsn.com
> Cc: Jan-Erik Soderholm <jan-erik....@telia.com>
> Subject: Re: [Info-vax] improving EDT
>

[snip...]

>
> A lot of fun stuff there...
>
> But, we are only talkning about giving (new) programmers on
> VMS a decent environment to work in. I do not understand why
> you drag clouds and all that stuff into the picture.
>

These were just IT examples of political correctness today
whereby some promote new stuff over old stuff just because the
new stuff appears to be "cooler" - not necessarily "better" than
existing capabilities.

> I do not understand your post. Are you saying that editing code
in
> EDT/TPU is "better" then editing code in one of todays popular
> IDE's?
>
>

It depends on the individual. Editors are simply another basic
tool for code development.

Fwiw, rather than jumping in with new editor developments, I
think there are already various editor options and capabilities
for OpenVMS that many here may not have really looked into yet.

Examples are eCubes NXTware (runs on different Windows, Linux,
MacOS clients, interfaces with NXTware server on OpenVMS and can
integrate with existing IDE's including CMS, has numerous
plug-ins available in Eclipse world),
http://www.ecubesystems.com/products/nxtware-remote.html (scroll
down and click on video testimonial)
http://www.slideshare.net/ecubemarketing/why-nxtware-remote-for-j
enkins (Jenkins integration)

jEdit:
http://www.jedit.org/
http://www.jedit.org/index.php?page=download

JED (I was not aware of until HB referenced)
http://www.jedsoft.org/jed/

LSE is perhaps not as cool from a multi-platform perspective, but
it is certainly a capable editor - especially if the primary dev
environment is OpenVMS (CMS, using OpenVMS native features etc.)
http://h20565.www2.hpe.com/hpsc/doc/public/display?docId=emr_na-c
04618749

[David Froble]

IanD wrote:
> Notepad++ comes with a huge number of programming language support
>
> It is one of the most commonly suggested editors in a number of the
> programming forums when people start out learning programming and ask what
> editor should they use
>
> One of the key strategic aspects of reviving OpenVMS was how to attract new
> talent, you will not do that asking people to learn any of the editors
> natively supplied with OpenVMS simply because new kids want portable skills,
> that means going with majority products and with ones they will be familiar
> with. After this, the framework can be embraced

Well, you do have a valid point. Just as I ask, what do I gain by dumping what
I know (EDT) for something else, those who know something else just as rightly
ask the same question. Why should they learn EDT (or whatever) when they
already know something that works for them?

[hb]

For JED: http://www.jedsoft.org/jed/
For additional JED modes: http://www.vms2linux.de/jedmodes.html, the MMK
mode: https://github.com/endlesssoftware/mmk/issues/89
For the Eclipse plugin: http://www.vms2linux.de/vms-plugins-eclipse.html

[Scott Dorsey]

Yes. But likely what YOU want is an integrated development environment that
continues to use EDT key sequences.

The IDE makes code debugging a whole lot easier, in part because you can
see more at the same time than you can looking through a little 80x24
windows. But there's no reason there shouldn't be an IDE with keymappings
and maybe some key commands for existing VMS users.
--scott

--
"C'est un Nagra. C'est suisse, et tres, tres precis."

[Arne Vajhøj]

On 11/17/2016 5:40 PM, johnwa...@yahoo.co.uk wrote:
> On Thursday, 17 November 2016 21:27:45 UTC, Jan-Erik Soderholm wrote:
>> But, we are only talkning about giving (new) programmers
>> on VMS a decent environment to work in. I do not understand
>> why you drag clouds and all that stuff into the picture.
>>
>> I do not understand your post. Are you saying that editing code in
>> EDT/TPU is "better" then editing code in one of todays popular IDE's?

>

> Better for who, in order to achieve what, exactly?
>
> Someone very familiar with VMS can be very productive on
> VMS, as is. What's the point of forcing folk like that to
> work unproductively in an unfamiliar IDE (especially in
> cases where VMS is still seen as a limited lifetime system)?

I don't think anyone is suggesting removing TPU or EDT from VMS.

Even though I suspect that a lot of those familiar with VMS
would be glad for the new tools if they spent some time
getting used to them.

> On the other hand there are lots of people out there who
> are familiar with stuff like Notepad++ or Eclipse or [etc]
> and it might not be a bad idea for someone (doesn't have
> to be VSI) to put some thought into making these folks life
> more comfortable and productive when working in and with
> the VSIVMS environment.

> More and better such stuff (or at least more visibility for
> decent stuff that exists) would probably be welcome.

Yep.

Arne

[Arne Vajhøj]

That will depend on some things:
- how much time you spend using an editor
- the complexity of the editing you do
- whether you work on other platforms as well
etc.

But let us randomly say that 50% of existing
VMS users and 100% of new VMS users would prefer
the new tools. Seems very relevant to me.

Arne

[Arne Vajhøj]

On 11/17/2016 6:03 PM, Scott Dorsey wrote:
> David Froble <da...@tsoft-inc.com> wrote:
>> Jan-Erik Soderholm wrote:
>>
>>> But, we are only talkning about giving (new) programmers
>>> on VMS a decent environment to work in. I do not understand
>>> why you drag clouds and all that stuff into the picture.
>>>
>>> I do not understand your post. Are you saying that editing code in
>>> EDT/TPU is "better" then editing code in one of todays popular IDE's?
>>
>> Ok, let me try explaining this way.
>>
>> Does any of these "new" "wizz-bang" methods have enough advantage for me to
>> forget using EDT, which I pretty much got down to "muscle memory", and spend
>> effort (don't know how much) learning something new?
>
> Yes. But likely what YOU want is an integrated development environment that
> continues to use EDT key sequences.

IDE for serious development.

But probably also a general purpose editor for casual development
and system manager tasks.

> The IDE makes code debugging a whole lot easier, in part because you can
> see more at the same time than you can looking through a little 80x24
> windows. But there's no reason there shouldn't be an IDE with keymappings
> and maybe some key commands for existing VMS users.

Most IDE's allow customizing key mappings.

But it will not solve the learning curve problem as a modern
IDE does not have a 1:1 mapping to EDT.

Arne

[Arne Vajhøj]

It is not just a matter of what one know.

The new tools have a lot more features than the old tools.

Whether those new features make it worth for you to switch to
the new tools when you know the old tools depends.

But practically the only reason to switch from new tools
to old tools with less features is if the new tools
are not available.

Arne

[Richard Maher]

On 18-Nov-16 2:58 AM, Hans Vlems wrote:
> I'll second that.
>
I agree but these days I use it about as much as vi

Oh hold on that me paraphrase the VMS crowd "What does change mode in
EDT give me that line-editing can't"

[Richard Maher]

On 16-Nov-16 11:35 PM, VAX...@SendSpamHere.ORG wrote:
> In article <F+hM4z...@eisner.encompasserve.org>, koe...@eisner.nospam.decuserve.org (Bob Koehler) writes:
>> In article <o0hkks$e45$1...@gioia.aioe.org>, Richard Maher <maher_rj...@hotmail.com> writes:
>>>
>>> For fuck sake how hard would it be to port Notepad++ to VMS?
>>>
>>
>> Why would anyone want to work on porting that worthless P.O.S.
>> to any platform?
>
> I believe Richard was being facetious.
>

Nope: - https://notepad-plus-plus.org/download/v7.2.html

[Simon Clubley]

On 2016-11-17, Jan-Erik Soderholm <jan-erik....@telia.com> wrote:
>
> A lot of fun stuff there...
>
> But, we are only talkning about giving (new) programmers
> on VMS a decent environment to work in. I do not understand
> why you drag clouds and all that stuff into the picture.
>
> I do not understand your post. Are you saying that editing code in
> EDT/TPU is "better" then editing code in one of todays popular IDE's?
>

Well hipsters these days think it's cool to use mechanical typewriters
in public[1]. :-(

Just tell them that using a character cell editor makes them look
cool and perhaps those editors will come back into fashion again.

[1] I am not joking; Google for hipster typewriters. One example:

http://www.telegraph.co.uk/men/the-filter/virals/11675023/Hipsters-really-are-using-typewriters-in-public-places.html

Simon.

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

[Simon Clubley]

On 2016-11-17, David Froble <da...@tsoft-inc.com> wrote:
>
> Does any of these "new" "wizz-bang" methods have enough advantage for me to
> forget using EDT, which I pretty much got down to "muscle memory", and spend
> effort (don't know how much) learning something new?
>

Use an editor which has the EDT keypad mappings but also has lots of
other features. This response is being typed in Emacs in character
cell mode on Linux but with a fully active EDT compatible keypad.

[Simon Clubley]

On 2016-11-17, Scott Dorsey <klu...@panix.com> wrote:
> David Froble <da...@tsoft-inc.com> wrote:
>>

>>Does any of these "new" "wizz-bang" methods have enough advantage for me to
>>forget using EDT, which I pretty much got down to "muscle memory", and spend
>>effort (don't know how much) learning something new?
>
> Yes. But likely what YOU want is an integrated development environment that
> continues to use EDT key sequences.
>

Emacs has this but as I understand it the EDT keypad support has been
removed from the very latest Emacs versions. Fortunately, I use the
LTS/stable versions of Linux so it's not a problem I will have to tackle
for several years.

[Kerry Main]

> -----Original Message-----
> From: Info-vax [mailto:info-vax...@rbnsn.com] On Behalf
> Of Arne Vajhøj via Info-vax
> Sent: 17-Nov-16 6:59 PM
> To: info...@rbnsn.com
> Cc: Arne Vajhøj <ar...@vajhoej.dk>
> Subject: Re: [Info-vax] improving EDT
>

Or it could be a case of older tools being simpler while at the
same time addressing 80+% of the actual work a developer needs to
do.

Some tools and technologies suffer from over-development whereby
more and more features get crammed into a product, rapidly
increasing complexity and in some cases, pricing. Many users
might not like all the new "limited value" features and might
prefer the older, simpler versions.

[Henry Crun]

+1
my fingers remember what to do in TPU/EDT/EVE, so much so that when I'm on a
non-VMS system there's a double effort - I have to switch off automatic actions
and reactions...

--
Mike R.
Home: http://alpha.mike-r.com/
QOTD: http://alpha.mike-r.com/qotd.php
No Micro$oft products were used in the URLs above, or in preparing this message.
Recommended reading: http://www.catb.org/~esr/faqs/smart-questions.html#before
and: http://alpha.mike-r.com/jargon/T/top-post.html
Missile address: N31.7624/E34.9691

[Craig A. Berry]

I resisted for some years the trend toward autocomplete, syntax
highlighting, on-the-fly type checking, standardized formatting,
applying the same edit to multiple lines at once, debugging within the
editor, etc. It is pretty hard to learn this stuff when you learned the
EDT keypad with your ABCs decades ago, but it's also hard to do without
once you make the effort.

Example: when you type

%INCLUDE "XYZ.INC"

in one of your BASIC programs, what if the editor put a red squiggle
under "XYZ.INC" if it couldn't find the file, letting you know
immediately that you had a typo? You wouldn't have to wait until you try
to compile it to see the problem. LSE was a step in the right direction,
but still separated compilation from editing.

So then multiply that times a hundred because there are dozens of
productivity boosters in a modern editor that the native VMS editors
don't do. Whether this is "enough advantage" for you personally is
something only you personally can determine. But marketing any system
today without these capabilities would just be embarrassing.

[Arne Vajhøj]

On 11/17/2016 8:53 PM, Kerry Main wrote:
>> Of Arne Vajhøj via Info-vax

>> On 11/17/2016 5:49 PM, David Froble wrote:
>>> Well, you do have a valid point. Just as I ask, what do I gain by
>>> dumping what I know (EDT) for something else, those who know something
>>> else just as rightly ask the same question. Why should they learn EDT
>>> (or whatever) when they already know something that works for them?
>>
>> It is not just a matter of what one know.
>>
>> The new tools have a lot more features than the old tools.
>>
>> Whether those new features make it worth for you to switch to
>> the new tools when you know the old tools depends.
>>
>> But practically the only reason to switch from new tools to old
>> tools with less features is if the new tools are not available.
>

> Or it could be a case of older tools being simpler while at the
> same time addressing 80+% of the actual work a developer needs to
> do.
>
> Some tools and technologies suffer from over-development whereby
> more and more features get crammed into a product, rapidly
> increasing complexity and in some cases, pricing. Many users
> might not like all the new "limited value" features and might
> prefer the older, simpler versions.

Most IDE's and editors today are free.

The hassle of IDE/editor complexity in the form of more
key shortcuts and more menu items and more available
plugins is negligible - one just don't use the featured
one don't need.

Arne

[David Froble]

Reasonable.

I've had many criticize me for not declaring all variables. My response has
been "that's the compiler's job, why should I do such work?". Nothing different
in having tools that attempt to do the grunt work for you.

I also do some Visual Basic work at times. I'm not so sure that I like it when
something that thinks it knows more of what I'm doing than I do causes
interruptions. For example, in the middle of typing something, I decide I need
to look elsewhere. The damn thing has to nag me about the incomplete verb or
whatever, and it is an interruption in what I'm concentrating on, and actually a
detriment rather than help. Some might roll eyes at this, but for me it's not
trivial.

I guess that what is helpful to some isn't necessarily helpful to others.

[Jan-Erik Soderholm]

Den 2016-11-17 kl. 23:40, skrev johnwa...@yahoo.co.uk:


>>>
>>>
>
> Better for who, in order to achieve what, exactly?

Better for someone new to VMS.
*I* can use EDT until retirement...

[Jan-Erik Soderholm]

Den 2016-11-17 kl. 23:43, skrev David Froble:
> Jan-Erik Soderholm wrote:
>
>> But, we are only talkning about giving (new) programmers
>> on VMS a decent environment to work in. I do not understand
>> why you drag clouds and all that stuff into the picture.
>>
>> I do not understand your post. Are you saying that editing code in
>> EDT/TPU is "better" then editing code in one of todays popular IDE's?
>
> Ok, let me try explaining this way.
>

> Does any of these "new" "wizz-bang" methods have enough advantage for me...

Are you a "new programmer on VMS"?

> to
> forget using EDT, which I pretty much got down to "muscle memory", and
> spend effort (don't know how much) learning something new?
>

Probably not. But I wasn't talkning about you (or myself, for that
matter).

[Jan-Erik Soderholm]

Den 2016-11-17 kl. 23:49, skrev David Froble:
> IanD wrote:
>> Notepad++ comes with a huge number of programming language support
>>
>> It is one of the most commonly suggested editors in a number of the
>> programming forums when people start out learning programming and ask what
>> editor should they use
>>
>> One of the key strategic aspects of reviving OpenVMS was how to attract new
>> talent, you will not do that asking people to learn any of the editors
>> natively supplied with OpenVMS simply because new kids want portable skills,
>> that means going with majority products and with ones they will be familiar
>> with. After this, the framework can be embraced
>

> Well, you do have a valid point. Just as I ask, what do I gain...

Probably nothing. But then, noone is talkning specificaly about you here.

And *I* will probably stick to EDT until retirement...

[Jan-Erik Soderholm]

Den 2016-11-17 kl. 23:43, skrev David Froble:

> Jan-Erik Soderholm wrote:
>
>> But, we are only talkning about giving (new) programmers
>> on VMS a decent environment to work in. I do not understand
>> why you drag clouds and all that stuff into the picture.
>>
>> I do not understand your post. Are you saying that editing code in
>> EDT/TPU is "better" then editing code in one of todays popular IDE's?
>
> Ok, let me try explaining this way.
>

> Does any of these "new" "wizz-bang" methods have enough advantage for me...

Another thing... I do not think that the new group that VSI is putting
together in Sweden to specifically look at new development tools, open
source and similar stuff, are specificaly targeted at you and me.
VSI are hiring new/your people right out of high-school for this.

I'm pretty sure this group is not looking at your specific needs, David.

You must let go with your self-centric view on VMS development and try
to look at the larger picture. Or don't, it doesn't matter much. :-)

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

Hiring new people straight out of school is good. Back in the day,
Sweden used to have a VMS community and some of them may still be
around and some of them may even have some worthwhile experience
of things non-VMS.

These folk, the ones that have been around a bit and seen the
reality as well as they hype, may be interesting people to have on
the team formally or informally, as they have experience of other
people's mistakes on both sides of the fence.

TLDR; Other people's mistakes are good things to learn from. Use
them well.

[Bob Koehler]

In article <b382429c-677c-4c89...@googlegroups.com>, IanD <iloveo...@gmail.com> writes:
> Notepad++ comes with a huge number of programming language support
>

> It is one of the most commonly suggested editors in a number of the program=
> ming forums when people start out learning programming and ask what editor =
> should they use

On what planet?

[Bob Koehler]

In article <o0lao9$7ij$3...@gioia.aioe.org>, =?UTF-8?Q?Arne_Vajh=c3=b8j?= <ar...@vajhoej.dk> writes:
> On 11/17/2016 9:06 AM, Bob Koehler wrote:
>> In article <o0iuam$kpd$2...@gioia.aioe.org>, =?UTF-8?Q?Arne_Vajh=c3=b8j?= <ar...@vajhoej.dk> writes:
>>> Notepad++ and Atom seems to be preferred general editors
>>> these days.
>>
>> Preferred by who? Nobody around here.
>
> Maybe not.
>
> But by the large developer community that exist outside
> of VMS.

Around here, there are only those who remember VMS. Except myself,
having it at home and knowing where it's still in use.

[Bill Gunshannon]

On 11/17/16 1:39 PM, David Froble wrote:
> Jan-Erik Soderholm wrote:

>> Den 2016-11-17 kl. 15:06, skrev Bob Koehler:
>>> In article <o0iuam$kpd$2...@gioia.aioe.org>,
>>> =?UTF-8?Q?Arne_Vajh=c3=b8j?= <ar...@vajhoej.dk> writes:
>>>>
>>>> Notepad++ and Atom seems to be preferred general editors
>>>> these days.
>>>
>>> Preferred by who? Nobody around here.
>>>
>>

>> Yes, and that is part of the general problem.
>> Many here still lives in the past. Doesn't work
>> as well today as it did in the past...
>>
>
> Don't know about that. I'm thinking that everything I do today is based
> on the past. As far as I can tell, things that have been around for a
> while still do everything they did when first developed.

Somethings they even do better than what is available today, but one
thing I have learned about our business is that old is always bad and
even if the path your on is headed for the end of a cliff you never
go back and try the other path that was probably the right one in the
first place.

>
> I have no problem adopting better methods, things, and such. It's what
> I've always done in the past, "the past" being everything up to each
> letter I type. But I will not adopt anything, just because it's new. It
> must show itself to be better, better enough for me to abandon what I
> already have learned.

Dave, it may scare you but there are definitely some things that we
share the same thoughts on.

bill

[Scott Dorsey]

In article <Lx0Ym3...@eisner.encompasserve.org>,

Unfortunately it is this one.

[David Froble]

Scott Dorsey wrote:
> In article <Lx0Ym3...@eisner.encompasserve.org>,
> Bob Koehler <koe...@eisner.nospam.decuserve.org> wrote:
>> In article <b382429c-677c-4c89...@googlegroups.com>, IanD <iloveo...@gmail.com> writes:
>>> Notepad++ comes with a huge number of programming language support
>>>
>>> It is one of the most commonly suggested editors in a number of the program=
>>> ming forums when people start out learning programming and ask what editor =
>>> should they use
>> On what planet?
>
> Unfortunately it is this one.
> --scott
>

Well, at least one thing is confirmed. Namely, I don't get out much, because
this thread is the first I've heard about it.

:-)

[Arne Vajhøj]

This one.

Arne

[Arne Vajhøj]

On 11/18/2016 2:37 AM, David Froble wrote:

> Craig A. Berry wrote:
>> I resisted for some years the trend toward autocomplete, syntax
>> highlighting, on-the-fly type checking, standardized formatting,
>> applying the same edit to multiple lines at once, debugging within the
>> editor, etc. It is pretty hard to learn this stuff when you learned the
>> EDT keypad with your ABCs decades ago, but it's also hard to do without
>> once you make the effort.
>>
>> Example: when you type
>>
>> %INCLUDE "XYZ.INC"
>>
>> in one of your BASIC programs, what if the editor put a red squiggle
>> under "XYZ.INC" if it couldn't find the file, letting you know
>> immediately that you had a typo? You wouldn't have to wait until you try
>> to compile it to see the problem. LSE was a step in the right direction,
>> but still separated compilation from editing.
>>
>> So then multiply that times a hundred because there are dozens of
>> productivity boosters in a modern editor that the native VMS editors
>> don't do. Whether this is "enough advantage" for you personally is
>> something only you personally can determine. But marketing any system
>> today without these capabilities would just be embarrassing.
>
> Reasonable.
>
> I've had many criticize me for not declaring all variables. My response
> has been "that's the compiler's job, why should I do such work?".
> Nothing different in having tools that attempt to do the grunt work for
> you.

Some languages encourage that - other do not.

> I also do some Visual Basic work at times. I'm not so sure that I like
> it when something that thinks it knows more of what I'm doing than I do
> causes interruptions. For example, in the middle of typing something, I
> decide I need to look elsewhere. The damn thing has to nag me about the
> incomplete verb or whatever, and it is an interruption in what I'm
> concentrating on, and actually a detriment rather than help. Some might
> roll eyes at this, but for me it's not trivial.
>
> I guess that what is helpful to some isn't necessarily helpful to others.

Most IDE's/editors today mark problems with the code and provide a way
to get the details of the problem. Insisting on something being fixed
right away is not common.

Arne

[Simon Clubley]

On 2016-11-18, David Froble <da...@tsoft-inc.com> wrote:
>
> I've had many criticize me for not declaring all variables. My response has
> been "that's the compiler's job, why should I do such work?". Nothing different
> in having tools that attempt to do the grunt work for you.
>

They are absolutely right to criticise you IMHO. Declaring variables
is protection against some types of silly typing errors and in more
modern traditional style languages this is a requirement (or you
get an undeclared variable error during compilation).

[Scott Dorsey]

Simon Clubley <clubley@remove_me.eisner.decus.org-Earth.UFP> wrote:
>On 2016-11-18, David Froble <da...@tsoft-inc.com> wrote:
>>
>> I've had many criticize me for not declaring all variables. My response has
>> been "that's the compiler's job, why should I do such work?". Nothing different
>> in having tools that attempt to do the grunt work for you.
>
>They are absolutely right to criticise you IMHO. Declaring variables
>is protection against some types of silly typing errors and in more
>modern traditional style languages this is a requirement (or you
>get an undeclared variable error during compilation).

Back in the 1960s there was a valid argument for automatic variable
assignment, and back then there were compiler listings and everyone
was warned to be very very careful to make sure that there weren't
any spurious variables showing up on their listings.

Today modern compilers and debugging practices that came into existence
in the 1970s and 1980s make listings far less important and one of those
things is forcing explicit declarations.

If you're using a language that will automatically set up undeclared
variables for you, it's just a matter of time before you accidentally
typo that comma and write:

DO 10 I=1.5

which is an assignment statement and not a loop statement. And if you
don't notice this new DO10I variable that shows up on your listing, you
are going to be in for a world of hurt.

[David Froble]

Simon Clubley wrote:
> On 2016-11-18, David Froble <da...@tsoft-inc.com> wrote:
>> I've had many criticize me for not declaring all variables. My response has
>> been "that's the compiler's job, why should I do such work?". Nothing different
>> in having tools that attempt to do the grunt work for you.
>>
>
> They are absolutely right to criticise you IMHO. Declaring variables
> is protection against some types of silly typing errors and in more
> modern traditional style languages this is a requirement (or you
> get an undeclared variable error during compilation).
>
> Simon.
>

So, who died, and left you god of variables?

It's worked for over 40 years for me.

YMMV

[David Froble]

Ya know, all you people telling me I need to do grunt work should also stop
telling me how great IDEs are and how they will save me time. Why is one so
good, and the other so horrible?

[Scott Dorsey]

David Froble <da...@tsoft-inc.com> wrote:
>Simon Clubley wrote:
>> On 2016-11-18, David Froble <da...@tsoft-inc.com> wrote:
>>> I've had many criticize me for not declaring all variables. My response has
>>> been "that's the compiler's job, why should I do such work?". Nothing different
>>> in having tools that attempt to do the grunt work for you.
>>
>> They are absolutely right to criticise you IMHO. Declaring variables
>> is protection against some types of silly typing errors and in more
>> modern traditional style languages this is a requirement (or you
>> get an undeclared variable error during compilation).
>

>So, who died, and left you god of variables?

Edsger Dijkstra.

[Scott Dorsey]

David Froble <da...@tsoft-inc.com> wrote:
>
>Ya know, all you people telling me I need to do grunt work should also stop
>telling me how great IDEs are and how they will save me time. Why is one so
>good, and the other so horrible?

The grunt work needs to be done.

You can do it, or you can let the IDE do it for you.

I suggest reading the Mythical Man-Month which explains a number of
practices you have advocated in the past and why they are problematic.
I fear you won't find any newer books even mentioning them because for
the most part programmers today don't even know what they are.

[Simon Clubley]

On 2016-11-19, David Froble <da...@tsoft-inc.com> wrote:
> Simon Clubley wrote:
>> On 2016-11-18, David Froble <da...@tsoft-inc.com> wrote:
>>> I've had many criticize me for not declaring all variables. My response has
>>> been "that's the compiler's job, why should I do such work?". Nothing different
>>> in having tools that attempt to do the grunt work for you.
>>>
>>
>> They are absolutely right to criticise you IMHO. Declaring variables
>> is protection against some types of silly typing errors and in more
>> modern traditional style languages this is a requirement (or you
>> get an undeclared variable error during compilation).
>>
>

> So, who died, and left you god of variables?
>

It's an opinion formed from exposure to a good range of programming
languages and environments.

BTW, have you ever tried a type safe language such as Ada or one of
the Wirth languages ? I'm assuming no on the former, but I don't know
about the latter.

Declaring variables is only the very first step towards safer code.

Even _C_ requires that you declare all your variables and that you
declare your intentions by declaring the types of the variables, at
least at a basic level. (However, languages like Ada take it to a far
higher level with the range of strictly enforced user defined types
you get there.)

> It's worked for over 40 years for me.
>

Unfortunately, what works for you doesn't work in other environments.

> YMMV

Very true.

[Chris]

On 11/17/16 22:22, Arne Vajhøj wrote:

>
> jEdit FTP and VMS also seems to work OK.
>
> Arne
>

I used to love using edt, but you do need the lk ketboard layout to
get the best from it. Notepad++ is pretty good for a modern editor,
but i'm still using nedit, which has been around for years and uses
the motif library, which vms may already have. Both notepad++ and
nedit support rectangular cut and paste, which I would find very
difficult not to have available now...

Chris

[Arne Vajhøj]

On 11/19/2016 3:05 PM, Simon Clubley wrote:
> On 2016-11-18, David Froble <da...@tsoft-inc.com> wrote:
>> I've had many criticize me for not declaring all variables. My response has
>> been "that's the compiler's job, why should I do such work?". Nothing different
>> in having tools that attempt to do the grunt work for you.
>
> They are absolutely right to criticise you IMHO. Declaring variables
> is protection against some types of silly typing errors and in more
> modern traditional style languages this is a requirement (or you
> get an undeclared variable error during compilation).

True.

But there are also some extremely popular languages that do not have
such requirement including PHP and JavaScript.

But I guess we can call them "non traditional".

:-)

Arne

[David Froble]

Now that alone has me thinking about looking at such an editor. Things such as
cutting a name and address out of a screen to make a shipping label would be
quite helpful.

[Arne Vajhøj]

On 11/19/2016 7:01 PM, Chris wrote:
> On 11/17/16 22:22, Arne Vajhøj wrote:
>> jEdit FTP and VMS also seems to work OK.
>

> I used to love using edt, but you do need the lk ketboard layout to
> get the best from it. Notepad++ is pretty good for a modern editor,
> but i'm still using nedit, which has been around for years and uses
> the motif library, which vms may already have. Both notepad++ and
> nedit support rectangular cut and paste, which I would find very
> difficult not to have available now...

jEdit also do rectangular. So does some EVE extensions.

Arne

[Arne Vajhøj]

On 11/19/2016 4:12 PM, Scott Dorsey wrote:
> Back in the 1960s there was a valid argument for automatic variable
> assignment, and back then there were compiler listings and everyone
> was warned to be very very careful to make sure that there weren't
> any spurious variables showing up on their listings.
>
> Today modern compilers and debugging practices that came into existence
> in the 1970s and 1980s make listings far less important and one of those
> things is forcing explicit declarations.
>
> If you're using a language that will automatically set up undeclared
> variables for you, it's just a matter of time before you accidentally
> typo that comma and write:
>
> DO 10 I=1.5
>
> which is an assignment statement and not a loop statement. And if you
> don't notice this new DO10I variable that shows up on your listing, you
> are going to be in for a world of hurt.

True.

But no modern language has the space not significant rule.

Arne

[Craig A. Berry]

On 11/19/16 6:55 PM, Arne Vajhøj wrote:
> On 11/19/2016 3:05 PM, Simon Clubley wrote:
>> On 2016-11-18, David Froble <da...@tsoft-inc.com> wrote:
>>> I've had many criticize me for not declaring all variables. My
>>> response has
>>> been "that's the compiler's job, why should I do such work?".
>>> Nothing different
>>> in having tools that attempt to do the grunt work for you.
>>
>> They are absolutely right to criticise you IMHO. Declaring variables
>> is protection against some types of silly typing errors and in more
>> modern traditional style languages this is a requirement (or you
>> get an undeclared variable error during compilation).

And anyone who cares about their code has been requesting the greatest
level of checking the given language offers for decades, from IMPLICIT
NONE in Fortran onwards.

> True.
>
> But there are also some extremely popular languages that do not have
> such requirement including PHP and JavaScript.
>
> But I guess we can call them "non traditional".
>
> :-)

PHP is probably a lost cause, but JavaScript is moving pretty rapidly
toward a number of modern features, and derivatives such as TypeScript
offer strong typing.

[Arne Vajhøj]

I believe it offers optional strong typing - like Basic.

Arne

[Hans Bachner]

Arne Vajhøj schrieb am 20.11.2016 um 01:55:
> [snip]

>
> But there are also some extremely popular languages that do not have
> such requirement including PHP and JavaScript.

As you mention JavaScript, I just recently stumbled over this link:

<http://jazcash.com/a-javascript-journey-with-only-six-characters/>

Interesting reading :-)

Have fun,
Hans.

[Chris]

On 11/20/16 00:57, David Froble wrote:

>
> Now that alone has me thinking about looking at such an editor. Things
> such as cutting a name and address out of a screen to make a shipping
> label would be quite helpful.

There are some things which are more convenient and made possible
within a graphical environment. Having said that, the only things I
normally have onscreen are 3 or 4 terminal sessions, editor and emmail
client.

Nedit has been around for years though and had all the features that
have since become standard expectations for a programmer's editor,
while still being fairly minimalist. It builds out of the box on any
unix or linux system i've tried it on and full sources are available,
so it may be not too much of a task to port to VMS...

In the end, choice of editor depends on what works for you. EDT cut
and paste was pretty good, but a gui environment seems more intuitive...

Chris

[Chris]

On 11/20/16 02:52, Arne Vajhøj wrote:

>
> jEdit also do rectangular. So does some EVE extensions.
>
> Arne
>
>

Last time I tried jedit, seemed a bit slow, but may be faster
these days. If you have java on vms, wouldn't jedit work right
out of the box, as it's written in java, fwir ?...

Chris

[Paul Sture]

On 2016-11-20, Chris <xxx.sys...@gfsys.co.uk> wrote:
> On 11/20/16 00:57, David Froble wrote:
>
>>
>> Now that alone has me thinking about looking at such an editor. Things
>> such as cutting a name and address out of a screen to make a shipping
>> label would be quite helpful.
>
> There are some things which are more convenient and made possible
> within a graphical environment. Having said that, the only things I
> normally have onscreen are 3 or 4 terminal sessions, editor and emmail
> client.

Lots of concurrent windows open here (on a Mac), and the GUI editors I
use allow me to open multiple files per window, and multiple windows.

At least one of those editors (non-free) allows the opening of files on
a project basis.

> Nedit has been around for years though and had all the features that
> have since become standard expectations for a programmer's editor,
> while still being fairly minimalist. It builds out of the box on any
> unix or linux system i've tried it on and full sources are available,
> so it may be not too much of a task to port to VMS...

There was a version of nedit floating around which worked on Alpha VMS
V7.3. A colleague who was then new to VMS (and still in his twenties)
wanted such an editor back in 2002.

> In the end, choice of editor depends on what works for you. EDT cut
> and paste was pretty good, but a gui environment seems more intuitive...

I still haven't found any other editor which allows you to move to the
start or end of the next/previous line like EDT or EVE with a single
keystroke. I really do miss that functionality.

--
You’ll find COBOL an easy language to learn and a FAR EASIER language to
master than many of the "modern" languages.
-- GNU COBOL Programmer’s Guide

[Arne Vajhøj]

On 11/20/2016 6:01 PM, Chris wrote:
> On 11/20/16 02:52, Arne Vajhøj wrote:
>> jEdit also do rectangular. So does some EVE extensions.
>

> Last time I tried jedit, seemed a bit slow, but may be faster
> these days. If you have java on vms, wouldn't jedit work right
> out of the box, as it's written in java, fwir ?...

If it is coded correctly and Java on VMS is good then yes.

But sometimes bad code make unhealthy assumption about
file/directory names or the format of text files.

And I believe the behavior of Java on VMS is very
dependent on the logicals controlling behavior
of various C API's.

So yes but.

:-)

Arne

[hb]

On 11/21/2016 12:25 AM, Paul Sture wrote:
> I still haven't found any other editor which allows you to move to the
> start or end of the next/previous line like EDT or EVE with a single
> keystroke. I really do miss that functionality.

If I have to use a text based editor I use JED (in Xterm, on VMS or on
Linux), which gives me exactly what you are looking for.

[Simon Clubley]

On 2016-11-19, David Froble <da...@tsoft-inc.com> wrote:

Built right into that machine in front of you in the form of EVE.

Doing the following in EVE on the above message:

1) "box select"
2) select the area
3) "box copy"
4) insert

gives this:

edt, but you
Notepad++ i
edit, which
ich vms may
ular cut and

[Simon Clubley]

I wouldn't hold up either of those languages as an example of how
programming languages should be implemented. :-)

[Bob Koehler]

In article <o0qf7m$k2l$1...@panix2.panix.com>, klu...@panix.com (Scott Dorsey) writes:
>
> Today modern compilers and debugging practices that came into existence
> in the 1970s and 1980s make listings far less important and one of those
> things is forcing explicit declarations.

Many of today's "modern" compilers simply won't generate a listing
for code that has some fairly common compile errors. Which makes
figuring out the error using listing features impossible, and can leave
you wandering around in a I tried A, I tried B, ..., action.

[Scott Dorsey]

I'm not sure what your argument is here. Modern compilers don't actually
generate listing files at all for the most part, because nobody really needs
them in the modern IDE world. It's probably been 20 years since I have seen
a listing file on a modern system.

What compiler are you using that generates a listing file but doesn't include
compile-time errors?

[Jan-Erik Soderholm]

I can do that in Putty no matter what editor i used.

[seasoned_geek]

On Wednesday, November 16, 2016 at 7:30:10 AM UTC-6, Henry Crun wrote:
>
> From my experience, everything (and lots more) that can be done with EDT can be
> done in TPU.
> E.g. I use TPU for text eding in 8-bit Hebrew (entering text right-to-left, the
> Learn function has no convenient equivalent in EDT, Spawn, etc. etc.
>
> To use in an Edit/Compile/Test cycle you could try LSE (also based on EVE) which
> has many language-specific constructs (though IIRC it costs...)

1) TPU has/had/may-still-have a horrible bug with respect to trailing whitespace. I ran into this decades ago when under contract at Nutrasweet R&D. Actually, one of their people ran into it and I got pulled in to help. I forget what the package was, I think it was an SQL/report generation tool a script was being edited for. When the file was edited with TPU feeding it into the application caused a horrific crash. After hours of trouble shooting and speaking with tech support (who could edit and run the exact same script with no problems) we stumbled on to tech support using EDT but this developer used TPU. A DUMP of the script showed the trailing white space.
Admittedly that particular parser had issues . . . but . . . I assume TPU may well now have a configuration/setting to auto trim trailing white space.

2) LSE is/was awesome. The downside is you used to have to purchase a DECSet license to obtain it. Yes, we should all be using CMS, MMS, and LSE but some places aren't.

[seasoned_geek]

On Wednesday, November 16, 2016 at 11:43:03 AM UTC-6, Bill Gunshannon wrote:
>
> Guess it depends on what you like in an editor. I've used
> these three and they are fine for simple editing tasks. I
> think MicroEmacs had a VMS version, but that would have been
> ages ago.
>
> bill

If EMACS had an editor it would be a great OS <Grin>

[Bill Gunshannon]

It's got a shrink. What more could you want?

bill

[seasoned_geek]

On Wednesday, November 16, 2016 at 6:40:57 PM UTC-6, Arne Vajhøj wrote:
>
> Yes.
>
> Notepad++ and Atom seems to be preferred general editors
> these days.
>
> Arne

Depends on what people are doing. On nearly every platform I've tried it on jEdit was slow with wretched font rending. I used it for quite a few years, but as you get older the eyes become less forgiving. Atom used to be awesome but the current release ported to Ubuntu for either 14 or 16 is bug ridden and prone to crashing.

When one is using the KDE desktop KATE is still pretty awesome. UltraEdit is nice but not free.

An editor which could/should port easily as well as being quasi user friendly would be the current iterations of NANO. It harkens back to WordStar days using Ctrl keys to do things. Been a very long time since I had a VT-220 style keyboard, https://upload.wikimedia.org/wikipedia/commons/9/9e/DEC-VT220-0a.jpg but one should be able to port the editor to use the Compose key for Ctrl if the driver will not do it natively.

For someone looking to build a terminal based IDE from scratch, they might wish to look at The Programmer's Workbench which came with MSC 6.0 and was heavily used for OS/2 development. http://www.os2museum.com/wp/os2-history/os2-1-2-and-1-3/msc6_pwb/ While it did support the mouse, you could navigate completely from the keyboard. Yes, it probably has oceans of DOS hooks and I must be violently ill recommending anything from Microsoft.

There was at one time a port of SCITE for VMS. It still has EDT key mapping files and is highly extensible if one wishes to learn LUA. http://www.scintilla.org/SciTEDownload.html

THE used REXX. I haven't used THE in years, but it did run in terminals. https://en.wikipedia.org/wiki/The_Hessling_Editor

I just hate VIM almost as much as I hate VI.

Since I started on the PDP I have used and still reach for EDT. LSE was great because it kept the EDT key mapping, but not everyone could afford the license.

Given today's yoot are mostly script-kiddies, unless they magically hit PF2, they would have no idea how to use it.

[seasoned_geek]

On Wednesday, November 16, 2016 at 7:26:51 PM UTC-6, Simon Clubley wrote:
> Just had a quick look.
>
> Notepad++ is indeed Windows only as you suggested in your other
> post and Atom appears to be GUI only; there is no apparent character
> cell support so you can forget about running multiple instances in
> multiple xterm/PuTTY windows or over a low bandwidth SSH connection
> where graphics rendering would take too long.
>
> Simon.
>
> --

notepadqq is the Linux port - but - it is still graphical