Has Mark Hammond attained pariah status yet?
Simon Brunning
>
> So asks an article on Mark in the IT section of today's Sydney Morning
> Herald - see
> http://www.it.mycareer.com.au/news/2002/03/12/FFXMFP3LOYC.html
Not with those of us who use the Win32all package which he has so generously
donated to the community, no. <wink>
I-know-you-were-joking-but-a-<wink>-wouldn't-have-been-out-of-place-ly y'rs,
Simon Brunning
sbru...@fuckmicrosoft.com
-----------------------------------------------------------------------
The information in this email is confidential and may be legally privileged.
It is intended solely for the addressee. Access to this email by anyone else
is unauthorised. If you are not the intended recipient, any disclosure,
copying, distribution, or any action taken or omitted to be taken in
reliance on it, is prohibited and may be unlawful. TriSystems Ltd. cannot
accept liability for statements made which are clearly the senders own.
Tim Churches
Herald - see
http://www.it.mycareer.com.au/news/2002/03/12/FFXMFP3LOYC.html
Personally I think that integrating Python into .NET is really
important. But try as I might, I just can't get too enthusiastic about
anything controlled by Microsoft, no matter how cool it is. I think a
photo which appeared in a recent issue of Scientific American sums it up
- see http://www.sciam.com/2002/0102issue/010250100.html
Tim C
Sydney, Australia
Robin Becker
Brunning <SBru...@trisystems.co.uk> writes
>> From: Tim Churches [SMTP:tc...@optushome.com.au]
>>
>> So asks an article on Mark in the IT section of today's Sydney Morning
>> Herald - see
>> http://www.it.mycareer.com.au/news/2002/03/12/FFXMFP3LOYC.html
>
>Not with those of us who use the Win32all package which he has so generously
>donated to the community, no. <wink>
>
>I-know-you-were-joking-but-a-<wink>-wouldn't-have-been-out-of-place-ly y'rs,
>Simon Brunning
>sbru...@fuckmicrosoft.com
all the evidence suggests that python.net (or whatever it will be
called) will not be python as we know it. Apparently .net will not
support multiple inheritance etc etc. There was an interesting recent
article on the subject of what .net/cli is not able to do. Another
example of embrace and strangle.
-If only Bill Gates had a mustache and lived in Baghdad-ly yrs-
Robin Becker
Daniel Dittmar
> called) will not be python as we know it. Apparently .net will not
> support multiple inheritance etc etc. There was an interesting recent
I don't see how the limitations of .NET (or rather the CLI part of it)
differ from those of the Java VM. So it might not be Python as we know it,
but it might be useful in certain applications.
And of course it might also be useful to port CPython to the .NET C compiler
and adding a .NET <-> Python layer.
> Another
> example of embrace and strangle.
.NET is more another failed example of UNCOL
Daniel
gbr...@cix.compulink.co.uk
> Another
> example of embrace and strangle.
Good pythonic behaviour, then.
Frithiof Andreas Jensen
by Microsoft - in essence they bar developers from using any knowledge
gained from access to source under license Microsoft to implement *any* Open
Source software, not "merely" software in competition with MS's
implementation.
Odds are that if one were to develop for .NET, one would be screwed later by
Microsofts legal machine for disclosing proprietary information and
violating the NDA ...
Samuele Pedroni
Tim Churches <tc...@optushome.com.au> wrote in message
mailman.101592644...@python.org...
> So asks an article on Mark in the IT section of today's Sydney Morning
> Herald - see
> http://www.it.mycareer.com.au/news/2002/03/12/FFXMFP3LOYC.html
>
It's probably just the journalist's fault but I find the article frivolous
and vague.
What he is concretely working about? What he contretely want to work about?
etc... There are many possible integration philosophies and technologies?
and both aspects tech/philosophy are far from irrelevant.
(The frozen Python.NET as a full fledged Python or limited
and altered, a new attempt in that direction, bridging ...)
Is he working in stealth-mode or it is just MS-kind PR <wink>?
What kind of Python technologies will the book treat?
Just some academic questions. regards.
Cameron Laird
My background is academic. I spend a lot of time nowadays
journalistically. I can assure you that this article is
*not* frivolous or vague--at least not in comparison with
most of what appears in the popular and trade presses.
My best to Mark, in any case.
--
Cameron Laird <Cam...@Lairds.com>
Business: http://www.Phaseit.net
Personal: http://starbase.neosoft.com/~claird/home.html
Samuele Pedroni
> In article <3c8e0...@news.bluewin.ch>,
> Samuele Pedroni <pedr...@bluewin.ch> wrote:
> >
> >Tim Churches <tc...@optushome.com.au> wrote in message
> >mailman.101592644...@python.org...
> >> So asks an article on Mark in the IT section of today's Sydney Morning
> >> Herald - see
> >> http://www.it.mycareer.com.au/news/2002/03/12/FFXMFP3LOYC.html
> >>
> >
> >It's probably just the journalist's fault but I find the article
frivolous
> >and vague.
> >
> >What he is concretely working about? What he contretely want to work
about?
> >etc... There are many possible integration philosophies and technologies?
> >and both aspects tech/philosophy are far from irrelevant.
> >(The frozen Python.NET as a full fledged Python or limited
> >and altered, a new attempt in that direction, bridging ...)
> >
> >Is he working in stealth-mode or it is just MS-kind PR <wink>?
> >
> >What kind of Python technologies will the book treat?
> >
> >Just some academic questions. regards.
> >
> >
>
> My background is academic. I spend a lot of time nowadays
> journalistically. I can assure you that this article is
> *not* frivolous or vague--at least not in comparison with
> most of what appears in the popular and trade presses.
>
What's your point, exactly?
My point was direct, maybe rude, and
there were too few <wink>s,
but basically I find that the article per se
does not do any good. I don't know
Mark personally and have nothing against
him.
I find its incipit quite frivolous, at least for the form.
YMMV.
Now it is *not* frivilous in the sense that it has
started, can start a "debate". But maybe
not the right way ...
And what datapoints we acquire through
the article?
That .NET is like Java and contour and not
simply Passport and Web Services.
I knew that.
That De Icaza is a misunderstood
guy: probably.
regards.
PS: - My background is academic too.
- Personally, I have nothing against Python and .NET integration,
but not all the possible technical solutions do the same
amount of good to all the "parties" involved.
A.M. Kuchling
> And what datapoints we acquire through
> the article?
Well, nothing much, which is all I'd expect from something so brief in
a general publication. Hopefully Mark will clarify at some point.
There does seem to be an increasing trend to vaporware announcements
about language unification, though. Witness Parrot, Mono, and
SmallScript [1], all of which talk about having support for many
languages including Python, but none of which have actually tried to
build that support or work with the Python community to build it.
--amk (www.amk.ca)
UTF-8 has a certain purity in that it equally annoys every nation, and
is nobody's default encoding.
-- Andy Robinson, 10 Apr 2000
Michael Chermside
people appear to be confused about what is and what isn't intended to be
taken with a <wink>. So let me state a few facts quite clearly, with
no sarcasm (or humor) intended.
1) Mark Hammond has made HUGE contributions to the Python community.
The degree to which he personally has contributed to usability on
the Windows platform is clearly an important contribution.
2) Mark generally has a decent sense of humor. So do lots of folks in
the Python community -- after all, it's not named for a snake! We
LIKE to laugh when journalists (or anyone!) makes a silly statement
and uses it as a lead.
3) Not everyone will want to use a .NET version of Python if Mark were
to create such a thing. (Not everyone will have a platform where
they COULD use it.) But everyone (pretty much everyone) will be
pleased that such a thing exists. Everyone (pretty much everyone) in
the Python community thinks that Python is a really cool tool and
would like as many people as possible to be able to use it! Python
everywhere!
4) Mark should have more support from within the Python community for
the work he is doing. So should Guido. And so does Andrew Kuchling,
Robin Dunn, Finn Bock, and MANY, many others. We could all use an
extra hand. I hope Mark gets the support he needs, much as I hope
all the other projects make headway.
Okay... now back to your regularly scheduled <wink> enabled sense of
humor. This message has been sponsored by the C.C.U.P.M.H.B.P.W.A.H.S.H.
(Committee to Cover Up the Pariah-ization of Mark Hammond By Pretending
We All Have a Sense of Humor), a sub-committee of the PSU.
Aahz
A.M. Kuchling <akuc...@mems-exchange.org> wrote:
>
>There does seem to be an increasing trend to vaporware announcements
>about language unification, though. Witness Parrot, Mono, and
>SmallScript [1], all of which talk about having support for many
>languages including Python, but none of which have actually tried to
>build that support or work with the Python community to build it.
That seems a tad harsh WRT Parrot. Early on they checked with
python-dev and got little in the way of interest, but they've been
keeping us informed irregularly. They did explicitly ask for help in
making sure that Parot would work with Python, and IIRC someone (Michael
Hudson?) tried doing a bit of work with Parrot/Python as a proof of
concept.
--
Aahz (aa...@pythoncraft.com) <*> http://www.pythoncraft.com/
The way to build large Python applications is to componentize and
loosely-couple the hell out of everything.
Martin v. Loewis
> And what datapoints we acquire through the article?
I think the article states a number of facts properly:
"Despite his best efforts, there just isn't a lot of interest from the
community in seeing Python move further into the Microsoft compound."
Certainly true; there is little interest in Python.NET in the Python
community (just as there is little interest in other branches, such as
Stackless, Psyco, and, to some degree, Jython).
"Hammond says Python developers tend to be wary and sceptical of
Microsoft's motives"
I'm quite sure Hammond said that, and I also agree with Mark. I'm
myself wary and sceptical of Microsoft motives, and I know many other
people who are.
"Hammond was working for Active State, a Canadian software tools
developer, but was laid off two months ago."
This fact is not news to us, but may be to other readers.
"One of the problems Hammond and de Icaza face is the morass of
confusion surrounding the exact nature of .NET. Microsoft may have
deliberately muddied the waters, confusing the casual IT user and
business managers especially."
That is certainly not overstated. I find confusion over what .NET
exactly is and does among my peers, and I agree that Microsoft itself
is source of some of the confusion.
Regards,
Martin
Neil Schemenauer
> IIRC someone (Michael Hudson?) tried doing a bit of work with
> Parrot/Python as a proof of concept.
That was AMK. It seems the Parrot guys are more interested in
micro-optimizations and adding regex support to the VM than in
supporting Python on Parrot. Too bad.
Neil
Samuele Pedroni
Martin v. Loewis <mar...@v.loewis.de> wrote in message
m3adtdb...@mira.informatik.hu-berlin.de...
> "Samuele Pedroni" <pedr...@bluewin.ch> writes:
>
> > And what datapoints we acquire through the article?
>
> I think the article states a number of facts properly:
>
properly? YMMV. You have restated and read them
properly <wink>.
regards.
Mark Hammond
> Um... look folks. We can all point at the article and laugh, but some
> people appear to be confused about what is and what isn't intended to be
> taken with a <wink>. So let me state a few facts quite clearly, with no
> sarcasm (or humor) intended.
Let me state first of all that I took the original post in the manner
which it was intended. From my POV, any publicity is good publicity.
To my mind, having a fairly large headline "Charmed by the Python" in
the 2 most respected newspapers in Australia (Sydney and Melbourne
broadsheets) is a nice coup. I certainly saw the humour in both the
newspaper article, and the newsgroup posts, so worry not.
The article was not 100% accurate. One 30 minute phone call, and this
article appears. The book I am working on now is *not* Python for .NET,
but just .NET. Python doesn't get much of a mention :(
And-what-is-wrong-with-being-a-pariah-anyway ly,
Mark.
David Ascher
I agree w/ Cameron's response. The article is remarkably technical for
a trade press article. If you want _technical_ questions, then I
suggest you ask Mark (politely, preferably -- accusing Mark of doing PR
is not going to get you anywhere =).
--david
Samuele Pedroni
David Ascher <Dav...@ActiveState.com> wrote in message
mailman.1015979203...@python.org...
>
> I agree w/ Cameron's response. The article is remarkably technical for
> a trade press article. If you want _technical_ questions, then I
> suggest you ask Mark (politely, preferably -- accusing Mark of doing PR
> is not going to get you anywhere =).
>
I find this comments quite out of target.
regards.
Samuele Pedroni
http://amywohl.weblogger.com/2002/03/04
The generally valuable point is the how-to one
can have personal resorts to get things straight,
e.g. a weblog.
regards.
Samuele Pedroni
I'm sorry. My issues were mainly with the article,
I'm aware that an amount of misquotation happens.
Mark Hammond <mham...@skippinet.com.au> wrote in message
3C8E7E90...@skippinet.com.au...
> From my POV, any publicity is good publicity.
I disagree with this POV, but YMMV.
You're playing with fire <wink>.
regards, Samuele Pedroni.
Sandy Norton
> Certainly true; there is little interest in Python.NET in the Python
> community (just as there is little interest in other branches, such as
> Stackless, Psyco, and, to some degree, Jython).
I'm somewhat bewildered by this statement. You are surely
underestimating the interest of the community in the above projects.
perhaps-the-subject-is-smiley-enough-ly-yrs,
Sandy
Martin v. Loewis
> Martin v. Loewis wrote in message:
> > Certainly true; there is little interest in Python.NET in the Python
> > community (just as there is little interest in other branches, such as
> > Stackless, Psyco, and, to some degree, Jython).
>
> I'm somewhat bewildered by this statement. You are surely
> underestimating the interest of the community in the above projects.
Maybe, but then, the community is not showing its interest very much.
I don't count articles in a newsgroup saying "this is a good thing",
"just continue with your efforts", "we definitely need this" as
interest. Instead, people contributing code, reporting bugs, or just
using the system in their applications show interest. This is the case
for Jython, but I don't think Armin or Christian got much actual code
contributions. Likewise, when Mark would not deliver a win32all build
for a few weeks, people would complain - but nobody would actually
step forward and publish a build herself.
Regards,
Martin
Max M
> I don't count articles in a newsgroup saying "this is a good thing",
> "just continue with your efforts", "we definitely need this" as
> interest. Instead, people contributing code, reporting bugs, or just
> using the system in their applications show interest. This is the case
> for Jython, but I don't think Armin or Christian got much actual code
> contributions. Likewise, when Mark would not deliver a win32all build
> for a few weeks, people would complain - but nobody would actually
> step forward and publish a build herself.
This could also just be a sign that there is interrest, but the people
using win32all might not be able to publish a build themself?
Not everybody is as knowledgable as Mark, Even though we work on the
superior Windows platform ;-)
regards Max M
A.M. Kuchling
> That was AMK. It seems the Parrot guys are more interested in
> micro-optimizations and adding regex support to the VM than in
> supporting Python on Parrot. Too bad.
To expand on that response a bit: you still can't use Parrot to
implement any vaguely "real" programming language because there's no
support for implementing variables. Once you've filled up all 32 PMC
registers, there's no way to spill any registers to memory because
there are no opcodes to do so. I'd really like to work on a
Python-to-Parrot translator some more, but right now there's no point
in trying to do so because there would be no way of implementing
Python's scoping.
Unfortunately development effort seems to be going to premature
optimization -- an x86 JIT, regex opcodes, counting the cycles taken
by the main loop -- instead of filling in the missing pieces, so I'm
getting increasingly skeptical that Parrot will provide a useful
substrate for anything.
--amk (www.amk.ca)
I suppose the best way to find out where you come from is to find out
where you're going, and then work backwards.
-- The Doctor, in "City of Death"
Gustavo Cordova
> Unfortunately development effort seems to be going to premature
> optimization -- an x86 JIT, regex opcodes, counting the cycles taken
> by the main loop -- instead of filling in the missing pieces, so I'm
> getting increasingly skeptical that Parrot will provide a useful
> substrate for anything.
>
> --amk
Hmmm... but it's gonna be a hell of a fast substrate, though.
It's like "going nowhere fast" :-)
-gustavo
Sandy Norton
> I don't count articles in a newsgroup saying "this is a good thing",
> "just continue with your efforts", "we definitely need this" as
> interest. Instead, people contributing code, reporting bugs, or just
> using the system in their applications show interest. This is the case
> for Jython, but I don't think Armin or Christian got much actual code
> contributions. Likewise, when Mark would not deliver a win32all build
> for a few weeks, people would complain - but nobody would actually
> step forward and publish a build herself.
Perhaps you're right, but the wonderful work Armin, Christian, and
Mark (God bless them) are doing qualify as deep magic in my book: it
just ain't easy to wrap your brain around the Psyco, Stackless, or
win32all C/C++ codebases. I'm sure many pythonistas would love to
contribute... maybe it's just a matter of time.
When it comes to use, however, there must be many instances which just
aren't talked about. Personally, win32all is an integral tool in my
python toolchest and I've used Psyco to more than double the speed of
one part of an application with great success. I haven't gotten round
to using Stackless to any great effect simply because my brain can't
accomodate first-class continuations... but I'm eagerly awaiting the
practical library that will accompany the new generation stackless:
microthreads are most welcome!
I think the best way the average non-guru level pythonista can
contribute is to document her/his experiences with various modules in
howtos, place examples in the pubic domain, and just spread the love.
all the best,
Sandy
Martin v. Loewis
> This could also just be a sign that there is interrest, but the people
> using win32all might not be able to publish a build themself?
>
> Not everybody is as knowledgable as Mark, Even though we work on the
> superior Windows platform ;-)
That is part of my complaint (or analysis), though: I certainly don't
expect that every user of Python or a Python port also is expert for
Python's internals. I would hope that a significant user community
would grow experts beyond the initial authors. This has happened for
Python, and this is also what Eric Raymond reports as the basis for
the "publish early, publish often" strategy. *Somebody* needs to get
involved, or contribute - else the author gets (rightfully, IMO) the
impression that he's left alone.
Regards,
Martin
Cliff Wells
Max M wrote:
> This could also just be a sign that there is interrest, but the people
> using win32all might not be able to publish a build themself?
>
> Not everybody is as knowledgable as Mark, Even though we work on the
> superior Windows platform ;-)
And quite possibly because not many people using the superior Windows
platform <cough> have a C compiler available to them. Probably due to
thoughtfulness on Microsoft's part ;)
--
Cliff Wells, Software Engineer
Logiplex Corporation (www.logiplex.net)
(503) 978-6726 x308 (800) 735-0555 x308
Gerhard Häring
> On Wed, 13 Mar 2002 09:35:48 +0000
> Max M wrote:
>
> > using win32all might not be able to publish a build themself?
> >
> > Not everybody is as knowledgable as Mark, Even though we work on the
> > superior Windows platform ;-)
>
> platform <cough> have a C compiler available to them. Probably due to
> thoughtfulness on Microsoft's part ;)
There are plenty of free C compilers for Windows: mingw, Borland,
lcc-win32. lcc-win32 even comes with an IDE.
If you dig long enough on the Microsoft site, you can (at least you
could at one time) even download a commandline version of the MS
compiler. AFAIK the commandline tools of the .NET SDK are also available
at no charge.
Apart from that, many people might fear that hacking the internals of a
COM implementation might make their head explode.
Gerhard
--
This sig powered by Python!
Außentemperatur in München: 14.5 °C Wind: 1.2 m/s
Cliff Wells
Gerhard Häring wrote:
> Apart from that, many people might fear that hacking the internals of a
> COM implementation might make their head explode.
Mine would. Come to think of it, having one's head explode might be
preferable.
lost-a-leg-just-trying-to-use-MS-Word-yesterday-ly yrs,
Peter Hansen
>
> There are plenty of free C compilers for Windows: mingw, Borland,
> lcc-win32. lcc-win32 even comes with an IDE.
>
> If you dig long enough on the Microsoft site, you can (at least you
> could at one time) even download a commandline version of the MS
> compiler.
As best I recall, that version is wholly unsuitable for real production
use. I believe it is the version without optimization (not the
"professional" edition) and the code it produces, when last I looked,
is _significantly_ larger and slower than the real version.
Okay for hacking, not for shipping.
-Peter
Patrick
> [...] The book I am working on now is *not* Python for .NET,
> but just .NET. Python doesn't get much of a mention :(
Mark, out of curiosity: how did you conclude that the Python community is
not interested in .NET compatibility? Was it based on the lack of
contributions to your trial implementation? Or was it a perception that
Python *users* are not interested?
Gerhard Häring
You're of course right. Has anybody tried their .NET SDK, yet. Does it contain
a usable commandline version of VC++7?
Gerhard
--
This sig powered by Python!
Außentemperatur in München: 8.8 °C Wind: 2.9 m/s
Neil Hodgson
> You're of course right. Has anybody tried their .NET SDK,
> yet. Does it contain a usable commandline version of VC++7?
The .NET SDK contains a usable compiler but as mentioned it does not
optimize. You may need to download extra include and library files such as
those provided by the Platform SDK which is also free. I already had these
from an earlier version of Visual C++. You may also need a version of STL
such as the free STLPort.
The code produced is not fast or small but works well enough for
applications that are not performance sensitive.
The mingw version of gcc and Borland's free C++ compiler are packaged
better than the free version of Microsoft's compiler and they include
optimization.
Neil
Stephen J. Turnbull
>> Not everybody is as knowledgable as Mark, Even though we work
>> on the superior Windows platform ;-)
Cliff> And quite possibly because not many people using the
Cliff> superior Windows platform <cough> have a C compiler
Cliff> available to them. Probably due to thoughtfulness on
Cliff> Microsoft's part ;)
Yeah, if there's a superior Windows platform around, they sure seem to
have forgotten to put it on the market. Beta test still?<wink>
--
Institute of Policy and Planning Sciences http://turnbull.sk.tsukuba.ac.jp
University of Tsukuba Tennodai 1-1-1 Tsukuba 305-8573 JAPAN
Don't ask how you can "do" free software business;
ask what your business can "do for" free software.
Mark Hammond
I would describe it more as a healthy skeptisism towards .NET, but I see
this starting to thaw slightly with the Open Source efforts. The Python
community is not overflowing with Microsoft apologists or zealots ;)
But if I combine this with the fact that new .NET beta releases often
breaks the Python compiler but I have only received 2 complaints in a
year, I concluded that interest is not yet high :)
Don't forget .NET 1.0 has only recently been released into production -
the "wait and see" attitude is quite prudent for many Python users.
Mark.
Brad Clements
> But if I combine this with the fact that new .NET beta releases often
> breaks the Python compiler but I have only received 2 complaints in a
> year, I concluded that interest is not yet high :)
>
> Mark.
>
And .NET is expensive to actually deploy, and we're all cheap.
-----= Posted via Newsfeeds.Com, Uncensored Usenet News =-----
http://www.newsfeeds.com - The #1 Newsgroup Service in the World!
Check out our new Unlimited Server. No Download or Time Limits!
-----== Over 80,000 Newsgroups - 19 Different Servers! ==-----
Delaney, Timothy
>
> "Mark Hammond" <mham...@skippinet.com.au> wrote in message
> news:3C907711...@skippinet.com.au...
>
> > But if I combine this with the fact that new .NET beta
> releases often
> > breaks the Python compiler but I have only received 2
> complaints in a
> > year, I concluded that interest is not yet high :)
> >
> > Mark.
> >
>
> And .NET is expensive to actually deploy, and we're all cheap.
After all, why else would we use a *free* language with no restrictions ...
;)
While I'm here, I'd like to thank Mark for all the work he has put into the
Python community and for his valiant attempt to date to corrupt .NET into
something possibly worth using (for customers, not me ;)
Tim Delaney