.NET and Python

[Van Gale]

"Ken Egervari" <ken at positive-edge.com> wrote in message
news:8nMm7.133116$n75.31959718 at news4.rdc1.on.home.com...
> So you would say Zope sites will run much faster than traditional
mod_python
> with apache sites?

I think it would depend on the programmer(s), which is the case with any
python application.  So, your next question might be, if it was the same
programmer doing both sites which would run faster?  In that case, I would
say there's still other factors, but I doubt either would be *much* faster
than the other.  You can build scalable high-traffic sites with either
approach, not to mention you might want to put either one behind distributed
squid servers anyway.

Side note: there are other approaches as well.  Speedy Aolserver with PyWX,
or WebWare, or several other Python web interfaces mentioned in this
newsgroup.

> I've looked into .NET and I hate to admit it, MS is
> become more right lately.

Microsoft marketing literature and hype has *always* been right.  For the
most part, Microsoft products eventually catch up to the marketing.

> Instead of crappy products, they are starting to
> produce stuff that does compete.

That's just wrong, they've always had at least *some* good products.  I've
always considered Word one of the best in that class (yeah... I liked
Interleaf and Framemaker better, but didn't have the huge bucks to buy
personal copies, so that made *them* uncompetitive).  Excel has always been
good.

> I think their framework is really good and would feel
> confortable implementing an 800 class system using their stuff.

Are there any live working sites using this framework?  I don't know, which
is why I'm asking, but frankly I would be surprised if there were.  It's
just too soon.

According to Microsoft marketing, IIS and NT have been stable, large
enterprise, systems since their release.  Most large companies believe this,
and then build huge server farms (and I do mean HUGE) so they can get the
appearance of stability.

My point is this... when you're talking Microsoft products don't believe it
until you see it.  The same holds true for any products of course, even open
source, but at least with open source it doesn't cost anything to "see it".

The .NET framework might rock someday, but do you want to implement your 800
class system now?  Or wait till .NET is past at least the initial stages of
maturity.

You can build your system now on top of Zope, or Apache+mod_python, or
Apache+WebWare, or Aolserver+PyWX, or a Java application server, and you're
working from a known solid foundation.  That's how I would look at it if I
wanted to sell an 800 class system as a reliable product.

On the other hand, if I wanted to market my personal services for $2000 a
day + expenses to people struggling to get .NET applications working, I'd
start by using that as my base.  You won't be able to sell your system, but
you'll still make a ton of money helping others work around the bugs and
pitfalls.

er... just my 2 cents :)