Richard,
Didn't mean to ruffle any feathers over at Gemtalk. Yes, Gemstone
probably can handle the problem. But it is about the only
IMPLEMENTATION that can! Yes, Gemstone/S is Industrial strength. But
Texas Instruments, Florida Power& Light, JP Morgan, ICE, NYK, COSCO,
are the ones that can afford to pay either $7000/core or $15000/core per
machine. (source: http://www.slideshare.net/esug/gemtalk ). (More than
I made last 3 years doing smalltalk !)
But GLASS is like your first FREE taste of crack. Once you get hooked
on it, you will have to rob little old ladies to afford the license fees
(or be a fortune 500 company). Looking at it this way, I would not do
my next startup on GLASS.
BTW, my opinions are based on 20 years of smalltalking. And BTW,
it is also my OPINION that Windows is not a suitable server platform.
Therefore, if it does not run on Linux, does not run near machine code
speeds and does not support multi-core out of the box, then I would not
choose it for a startup's web application development. There are so
many other good choices out there
For more about how I formed my OPINION on windows look here:
http://redmondmag.com/articles/2013/08/22/windows-8-security-issues.aspx
Post by Richard SargentYou suggest a proprietary implementation is some kind of evil.
Yes, I have had a lawsuit threatened by Getty Images for a tiny image
that was on a website that I bought from a guy, that was developed in
eastern europe. Copyright laws can kill you. Why take the chance that
some software vendor is going to come in and shut you down? Did you
know that by US copyright law, the plaintiff is allowed to recover
attorneys fees from the defendant? A proprietary software company can
shut you down just for the fun of it by burying you in legal fees! My
OPINION: Start your company right, avoid proprietary software platforms
at all costs. Again, just an opinion based on experience (and $1000
license fee for a tiny website image).
Post by Richard Sargent"That is why the language is dead" is pretty simplistic."
Yes, and I stand by that overly simplistic statement. Think about it.
Look around at all the smalltalk implementations and tell me why I
cannot run multiple cores at machine code speeds? There is one guy on
Pharo who is trying, but one guy cannot fix a dying industry.
But, every time you turn around somebody is building a NEW smalltalk
dialect. Whoop-Dee-Do! Every smalltalker wants to play in their own
little sandbox and every company wants to charge an arm and leg for
their 30 year old implementation. The excuse is always the same: we
need to feed our hungry, hungry programmers and just think of all the
support you get! Net result, everybody is pulling in different
directions and smalltalk keeps dying.
Also, every few months another debate breaks out in this newsgroup about
why smalltalk is dying and nobody can agree on anything. Everybody
keeps arguing about it and everybody digs in their heels.
I stand by my OPINION that too many proprietary dialects caused the
downfall of smalltalk.
Post by Richard Sargentsome of which is rubbish, and some of which is just plain uninformed
and erroneous
No, It may be a lot of opinions and it may have made you angry, but that
is not the same as erroneous rubbish. My opinions are based on personal
experiences and conclusions drawn against the facts. They are based on
30 years of programming experience and 20 years with smalltalk.
(oh, and did I mention 15 years of debugging windows viruses for
everyone in my extended family). But show me some new facts and I might
change my conclusions.
No hard feelings though, if I ever get up to Beaverton, I'll buy you a
beer.
Meanwhile, as for the Führer, he is not really looking for a solution
anyway. So don't sweat it.