Phoenix.pm: SCO and EFF -- do something

Scott Walters scott at illogics.org
Wed Sep 3 10:25:58 CDT 2003


I don't remember posting anti-Microsoft rants to the list. I think you must
be thinking of an actual Perl Mongers meeting =)

Well, let's see. On the Perl 6 internals list, the discussion of "why do we
need keyed varients for every access method" has cropped up yet
again. Dan is getting better and better at smacking that down and everyone
on the list for any period of time is really good at ignoring the whole
thing. The last Apocolypse (6) is extremely demanding of the language
side and it will take some time for the language people to catch up.
The Perl 6 grammar is useful for some simple things. The main drawback
is lack of anything like CPAN (yes, this is a more serious liability
than lack of objects). Hence PONIE was introduced, to make P5 run under
P6 by porting the guts of P5. Wouldn't it be easier to write a Parrot
gcc backend and automate the porting of the P5? I don't know. Perl
Design Patterns prompted an interesting exchange with Apress. For
those of you who don't know, I've been trying to sell Perl Design Patterns
to a publisher for some time. I've essentially ran the gauntlet of publishers.
I have 3 or 4 left to tap. The one thing I've learned is going through normal
channels does not work for someone without a recognizable name like myself.
I got Apress attention for a while because it turns out they had already
signed a deal with someone else (who IMHO is butchering the idea). 
There was some discussion of different ideas that could be done. They
fainted lack of interest in the subject, not wanting to leak that they
were working on it and as soon as it was announced, stopped returning
my emails. I'm feeling a little abused after that. We had been
seriously discussion doing something called "Perl 6 NOW", where
ideas slated to be introduced in the Perl language in Perl 6 would
be demonstrated in Perl 5 in a back-ported form. There are a huge
number of these - from the Coro module (giving you coroutines
and continuations in Perl) to object oriented error handling to
switch, the Perl6::* tree on CPAN, PDL, Class::Multimethods, Next.pm, 
threads::shared, Concurrent::Object for OO thread interface, Sub::Lexical for 
lexically scoped subroutines, Quantum::Suppositions for any() and all() like 
behavior, and some other stuff. I thought it was rather brilliant. They're
probably paying someone else to do it now using my outline and proposal.
Bastards. So I'm thinking of doing a free version and trying to beat
them to the punch. And taking on another major documentation project.
Or maybe it is retarded and I just haven't realized this yet. The
last major thing I was thinking of biting off someone beat me to -
Coro.pm. If you aren't familiar with the idea of a coroutine (which
will be in Perl 6 core), it goes something like this: a subroutine
can return, but unreturn later. It's lexical context is kept in
tact. Of course, "return" and "unreturn" aren't the keywords used.
That would be strange. Threads are complex and race conditions
and dataflow between threads confuse the heck out of people. Very
few people can code threads effectively. Coroutines provide a similar
service but tend to be more easily grasped by novices and more
easily handled by experts. It is a kind of cooperative multitasking,
you might say. Very handy for things like select() loops that
wait for activity on a number of filehandles. Things like POE try 
to solve this, but your handler is called and has to finish
executing and return, so it can't do things like store context
in variables on the stack. It can't have any stack at all. It
has to return. Hence, Coro is a far more powerful replacement
for POE and comes with the infrastructure you'd expect. Very nifty.
As I said, I was just starting to play with the idea of doing something
similar. I resisted the urge to chronical my communications with 
publishers for a long time, as I've not seen other people talk
about these rather senstive dealings, but I've essentially lost
the battle at this point and I'm feeling rather pent up, so I'm dropping
the facade. And I want to get this off my chest. This has been a huge
source of frustration and anxiety. My first serious contact was GNU Press.
They were pleasent, responsive, and encouraging, but unable to offer
anything in the way of editorial guidance, though this may have changed.
I'm due to talk to them again. If I do and something happens, then
the shroud of secerecy wouldn't be needed. As noted, I talked to Apress
for a while. Towards the tail end of that, I was talking to Nathan
Torkington for a while. You might remember him from the Perl Cookbook.
ORA made him the security editor, and I had the pleasure of meeting him
at DEFCON. I dropped him an email mentioning my mission after the
fact and managed two email turnarounds before he stopped responding.
ORA had previously completely ignored the proposal. Since it takes
a day to put a proposal according to a publishers template and spin
everything for them, not hearing back with even a "thanks but not
thanks" is a real slap to the face. This is exactly the treatment
I got from about 20 publishers. And yes, I spell check those ;)
Reguardless, Nathan is a very cool guy, and I have no doubt whatsoever that he 
is just painfully busy. I'll have to start thinking about doing Perl conferances.
On the other hand, I hate bugging busy people with my petty concerns.
Trying to talk to publishers makes you feel very small and very stupid,
and I'm feeling very grateful for Phoenix Perl Mongers through this
whole thing, for not making me feel small and stupid, and being 
in fact a receptive encouraging resource. Between that and
client abuse overload, I've taken a mental vacation the past
few days. I'm watching TV and I hate TV. It is *that* bad. When
I come back, I expect I'll do one of a few things. Do another
iteration of typesafety.pm, write something on Perl Assembly,
do the Perl 6 NOW thing for free and post it, talk to GNU Press
again, or write a Perl 5/B backend for the Perl 6 compiler.
The last would be an excellent hack, IMHO. It would solve the
Ponie problem since Perl 6 would be running on the Perl 5 VM
along with anything else from CPAN. Some things would have to
be done in software - multiple dispatch for instance. It would
also be ironic. The primary point to Perl 6 is to get rid of 
the nasty old Perl 5 guts. People don't care about the guts,
and IMHO, they should be refactored and cleaned up, not rewritten.
Rewritting large software programs is a disaster every time.
Remember Netscape 5? Of course you don't. It doesn't exist. Or
wasn't released or finished. Netscape worked for years trying to
rewrite their browser from scratch while Microsoft eat their
lunch. People aren't interested in which VM the language runs
on, but they are interested in the new language features, so
getting the experimental Perl 6 grammar to generate bytecode
for the Perl 5 VM would be very cool and very practical. And it would
piss a lot of people off. Which is usually my real goal. typesafety.pm
pisses off the Python people - giving Perl true type safety while
Python, supposedly all OO and stuff, goes without (bwahahaha!). 
http://perldesignpatterns.com got a facelift thanks to a volunteer
off of the Internet. I added something where you can comment
on pages without having to edit them. TinyWiki is still far
more popular Perl Design Patterns. Considering the code, I think
this says something: http://perldesignpatterns.com/?self -
code to be ashamed of. Or proud of. Depending on your culture.
Perl culture is pathological. Tell people that they can cleanly
write large programs in Perl and they scoff. Write cryptic
gibberish and they worship you. This excellent article
was posted to PerlMonks: http://perlmonks.org/index.pl?node_id=285065
I'm scrottie on there, by the way. Simon, who posted that, is the
one who has been helping with the Wiki. I'm trying to get 
content out of him but all he wants to do is hack on the engine
and ActiveWikiPages code. From all of this - the downplaying
of patterns, the hostility, the confusion, and outright
resentment, I've desided that Patterns should be dropped
from the name. "perldesign.com" is taken, however. Perl programmers
aren't openly hostile to software engineering concepts and 
discussion of design - their eyes only glaze over. I mentioned
some time back that my resume ranked very highly on
Google for search terms such as "computer programmer resume" yet
I got no jobs from it. A lot of prank pages, but no jobs. Curious
what was putting people off, I put a survey thing on it. The number
one survey item is "Dislike Perl". Perl programmers actively
dislike Patterns and the world actively dislikes Perl. Fitting,
perhaps. That is pretty much everyone on my mind Perl related.
Your turn to pick up the conversation. I wish I could do
something other than bitch, but I'm pretty fed up and burnt out
right now. Atleast griping about Microsoft is generally amusing.
Not to presume that everyone on this list cares what I've
been up to, but if anyone has related thoughts or opinions
it might be fuel for Perl related discussion.

Best wishes,
-scott


On  0, Tom Achtenberg <TomA at fh.org> wrote:
> 
> OK, I'll come right out and say it.  I do not want to see political discussions on this list.  Since Perl is cross platform anti Microsoft rants don't belong here either.  There are plenty of other lists out there if you need to forward petitions and whine about Microsoft.  Lets keep this one to the topic at hand, Perl.
> 
> -----Original Message-----
> From: Scott Walters [mailto:scott at illogics.org]
> Sent: Tuesday, September 02, 2003 10:48 PM
> To: Tom Achtenberg
> Cc: phoenix-pm-list at happyfunball.pm.org
> Subject: Re: Phoenix.pm: SCO and EFF -- do something
> 
> 
> Perl. That makes this discussion off-topic.
> 
> Atleast mostly off-topic. The state of Free Software legally and politically
> is interesting to people involved in the gospel and development of Perl.
> 
> Like any good off-topic discussion, it is being kept short. We've had some
> bad off-topic dicussions on here (programming licensing went on and on).
> 
> Now, there is a chance you meant to imply you wanted to see less off-topic
> discussion and you weren't merely asking if this was considered on-topic.
> If that is the case, I don't think anyone will be offended if you come
> right out and say that.
> 
> Best wishes,
> -scott
> 
> On  0, Tom Achtenberg <TomA at fh.org> wrote:
> > 
> > Is this a Perl discussion list or a political discussion list?
> > 
> > -----Original Message-----
> > From: Elston, Jeremy [mailto:Jeremy.Elston at schwab.com]
> > Sent: Tuesday, September 02, 2003 8:53 AM
> > To: 'phoenix-pm-list at happyfunball.pm.org'
> > Subject: RE: Phoenix.pm: SCO and EFF -- do something
> > 
> > 
> > Gotta love those Germans.  They sure know how to take care of things
> > quickly!
> > 
> > http://www.heise.de/newsticker/data/anw-02.09.03-000/
> > 
> > 
> > -----Original Message-----
> > From: johnb [mailto:johngnub at cox.net] 
> > Sent: Saturday, August 30, 2003 5:14 PM
> > To: phoenix-pm-list at happyfunball.pm.org
> > Subject: Re: Phoenix.pm: SCO and EFF -- do something
> > 
> > 
> > DONE, and passed it as as well;
> > 
> > JB
> > 
> > On Friday, August 29, 2003, at 12:07 PM, Scott Walters wrote:
> > 
> > >
> > > Hi,
> > >
> > > http://action.eff.org/postcard/index.asp?2003082995488702
> > >
> > > EFF is doing a letter writing, fax, emai, campaign to congress to make
> > > them
> > > aware of SCO's extortion. A little pressure can sick an attorney 
> > > general
> > > on SCO, and as you know, the attorney general is something of a 
> > > dedicated
> > > public prosectuor with an an office full of lawyers making sure 
> > > companies
> > > follow the law too and they act primarily in response to consumer 
> > > complaints.
> > > Fraud, false advertising, and extortion are all on their list. But I 
> > > rant.
> > >
> > > So, anyway, if you happen to think that sueing people for money to 
> > > bolster their case against IBM while refusing to provide any evidence 
> > > at all for people to make up their own mind with even if you did want
> > > to settle your debts is fraud, send a fax.
> > >
> > > -scott
> > >



More information about the Phoenix-pm mailing list