Re: FN-FORUM: What's the best development platform for creating Web 2.0 business applications?

date posted 19th November 2007 22:02

Gary Short wrote:


>>> On 19 Nov 2007 13:08:13 -0000, Carrie wrote
>>>> First I wouldn't use a Micro$oft server - unless you paid me.
>>>
>>> I'd be interested to know if that was for philosophical or technical
>>> reasons and if it was for technical reasons, what those reason were.
>>>
>>
>> Both really, I guess it would be my early experience with IIS sever,
>> which was a mess until a while after 2000 (the year)
>
> Ah cool, so not based on the current product set then?

No I have done some work with .NET and C# a couple of years ago,
but I wasn't very impressed - better yes - but still rather emcumbered,
and obfusticated. No one has asked for it since.

>
> No, scaleability isn't about going faster. Its the throughput increasing
> as the load increases; not really anything to do with execution speed.
>
yes
but surely one should use the machine available to its best effect, at
least just in terms of green issues.
as I said, it is about the appropriateness of the solution, by
separating different functions I can steer the load, and anticipate it.
This is working engineering rather than Rolls Royce or Aston Martin,
think Peogout 306 (Dturbo - actually) it can't go as fast as the Aston,
it is not as well engineered as the Rolls, but in general it can still
cope with the demands of prolonged motorway driving.
And doesn't cost nearly as much to buy or run!
As you would define it real scalability has honestly never been a
serious issue.

>
> Just out of interest, what world do you think I inhabit?
>

One where you have the time to discuss academic questions like this!

Yes I agree - in an ideal world where we got to to the absolutley best
job we possibly could, and had generous development time it would
allways be best to build in scalability, and elegance, and re-useailty,
and maitainability.

Hang on - in an otherwise diposable society where a new technology will
be available next week - why bother?

In a freelancer world you usually only get asked to update something
long after the last person who did it left, at which point the
re-usability of the original code is totally irrelavant, even if they
did have the faintest clue.

In practice, in many years of freelance (and employed) programming,
none of that has ever been relavant, or likely to win the business.

I leave my code as I would like to find it, /*commented */, concise, and
working.
Mostly so if I have to go back later on I can remember what I was doing
without trawling through my old log books. It works for me.

If something takes off, then I can charge them to addresss scalability
issues. Depending upon their particular requirements.
I wish! - I suspect that it may be a once in a lifetime surprise!

Sorry if that is a little down to earth for you.

Just out of interest how many years have you been in the business?


Carrie