Python in the Mozilla world

[Eric S. Johansson]

Steve Howell wrote:
> --- "Eric S. Johansson" <esj at harvee.org> wrote:
> 
>> Steve Howell wrote:
>>> --- "Eric S. Johansson" <esj at harvee.org> wrote:
>>>
> http://www.voidspace.org.uk/python/weblog/arch_d7_2007_04_28.shtml#e702
>>>> interesting.  Very interesting but I suspect the
>>>> message is "don't hold your 
>>>> breath but don't give up hope."
>>>>
>>> Exactly. :)
>> This is one of those things where a foundation, with
>> a smart person for making 
>> really simple things should get some money from the
>> foundation and get it built. 
> 
> Well, this is a case where smart people (including Jim
> Hugunin) are getting some money from Microsoft, which
> probably still has more dollars to spend than the
> average foundation.  And some really smart people in
> the open source world (Miguel de Icaza, etc.) are
> tracking to it, according to the article.

this is good to know except I am somewhat cautious about the end result given 
that it's from Microsoft (explained below)

   As far as
> "really simple things" goes, it doesn't really fit in
> that category.

I spent way too many years as a designer of software and then systems and may 
have found what may be a universal constant based on how people understand 
complexity and politics.

In the beginning, almost every idea is simple.  simplicity helps you get a rough 
go/no go estimation of project success.  Simplicity helps you communicate the 
essential elements of the project to others so that you can get them to buy in 
(funding, resources).  However as you develop use cases, the project gets more 
and more complex those in command issue to implement at the worst possible time 
because they are nervous about the increased complexity blowing up the budget.

The really cool thing about complex systems is that it's like climbing a 
mountain.  You never know when you reach the top until you are almost there. 
the simplicity doesn't emerge until you're almost done but it will emerge.  If 
the apparent solution keeps getting more and more complex, then you really don't 
understand the problem and shouldn't even try to implement the solution.

this is a pattern I can almost take to the bank.  Simple idea, complex partial 
solution, simpler solution.  only experience will tell you when you have gotten 
the most simplicity out of the solution possible.  Now granted, some problems 
are inherently complex but if you contrast mid-understanding versus final 
solution, there will always be a big difference and complexity.

I've always thought Microsoft interfaces were horribly horribly complex and I 
think it's due to two factors.  The first is that they implement at the worst 
point in their understanding, and it's a barrier to entry for all third-party 
developers.

If you don't agree with these ideas, that's cool.  We can agree to disagree but 
this has been my experience over many years.

> As much as I love Python, though, I still think the
> future of browser-side programming for the next five
> or ten years is Javascript, and I think the real
> productivity gains are going to come when Javascript
> libraries mature, it becomes more common knowledge
> which libraries are best of breed, etc.

that's really a pity.  Although that is consistent with a recent announcement I 
thought regarding IBM and their funding of the Linux desktop accessibility.  I 
personally hate the whole Ajax JavaScript crap because it is really inaccessible 
on multiple levels.  a sardonic corner of my mind wonders if the productivity 
gains you mention it will be offset by the productivity losses caused by the 
unbelievably horrible debuggers and IDs available and users trying to use user 
applications with even worse user interfaces than we have today.  But I will 
say, they will be prettier with all sorts of flashing colors and moving things. 
  Skins!  Don't forget the choice of skins you'll have for applications just by 
changing a cascading style sheet.  Who needs usability when you can the color 
and type font of your user interface to match your whim...grumble

I think your observation is far more correct than either of us want it to be. 
This is yet another good reason why folks should really push for a simple 
browser environment to replace JavaScript.  I really don't think it's horribly 
hard but on the other hand, I don't understand the full environment (starting 
with a simple idea... :-)

Thanks for the post.  It's triggered some nice ideas in my head.