|
|
Posted by Harry Slaughter on 07/25/05 01:37
Jerry Stuckle wrote:
>
> Sure I do. I have almost 40 years of programming experience.
>
> The point here is - if you're not sure you can do it in the timeframe
> given, you can't. You obviously are relatively new to this, and a bit
> unsure as to what you're doing.
I've been doing Internet development for about 10 years, but I've never
been responsible for developing an entire product from scratch with no
dev team and little to no specification and a timeframe measured in
weeks. I know most experienced developers wouldn't touch something like
this. But I'd like to make an effort to match the client's expectations
with the resource available for the task, me.
I'm aware of what happens when you try to develop before you know what
you're developing. And I can recognize mutually exclusive demands (the
old "good, fast, cheap, pick any two" rule). But before I simply bail on
the task, I want to try to find a way to make it work.
Since there's no time for real design and working without one would be a
mistake, I think my only option here is some form of framework that will
allow me to quickly put together a basic website with all the generic
features the client is looking for (content creation, user comments,
forums, user auth/session, site layout/design management...), while
freeing me up to develop the unique features of the website.
There's no shortage of options for this type of framework
(http://www.cmsmatrix.org/). Many of these tools are very feature rich.
Most of them allow you to write and incorporate your own tools into
their framework.
The only requirement not generally addressed by these tools is
scalability. The one unanswered question is how much load can these
frameworks support (concurrent users and/or [typical] page requests per
timeframe)?
Drupal, for example, is used by Mozilla.org, which I'd guestimate would
be somewhere in the top 20 percentile of high traffic websites. Of
course, mozilla.org does not utilize a lot of the features of drupal and
perhaps is not the ideal measure of drupal's scalability. Nonetheless,
drupal is obviously capable of scaling well past the needs of the
average website.
But what if the client wants to be prepared to support hundreds of
thousands of concurrent users? Yes, I know what you're all thinking :)
That's quite an ambitious target. And it's not one that stands a
statistically significant chance of ever being reached.
My thought is that something like drupal would be capable of supporting
a site that is popular enough to have proven or disproven its business
plan. At that point, it may or may not be necessary to redesign the
site. But also at that point, there would be the money available to hire
a complete engineering team to address problems.
Yes, the requirements are ill-defined and the expectations unrealistic.
But I think a single developer with the right framework could stand a
chance at developing a decent, customized website in a relatively short
period of time. A site which would scale at least to the point that the
site is somewhat successful and capable of generating enough money to
take it to the next level, whatever that may require.
Ok, I've lost myself. :)
Navigation:
[Reply to this message]
|