Ken at adOpenStatic.com
Tue Oct 30 08:15:30 CDT 2007
From: thelist-bounces at lists.evolt.org [mailto:thelist-bounces at lists.evolt.org] On Behalf Of Shawn K. Quinn
Sent: Tuesday, 30 October 2007 5:09 PM
To: thelist at lists.evolt.org
> ----- 8< -----
> As bandwidth increases, a compelling SDK should emerge that will provide
> developers with a standard interface for online apps
> ----- >8 -----
> Full story here:
> With regard to programmers giving up on optimization: this is exactly
> why we are in the sorry state of affairs we are in right now.
> Programmers are counting on new PCs to be twice as fast every couple of
> years and keep adding more and more bloat to keep up instead of
Some people need to solve real problems - and real people cost money. So the problem solving needs to cost less than the cost of the problem in the first place.
And what you might "optimisations", other people might call "a pointless waste of time". Optimisation is like "security" - a potentially bottomless pit of developer time and expense.
Instead, determine what your business requirements are for response/processing time, hire good developers and PMs, and you'll deliver a project that meets specifications, and on budget.
By having higher level tools (rather than writing email clients in ASM), we can have far greater developer productivity per hour (or dollar spent), and that means we solve far bigger/greater problems that would simply by insurmountable with the tools we had before.
We sacrifice, perhaps, some performance potentially, but unless that's a business requirement, then it's a moot point. Some things (like video card drivers) need a very high level of optimisation. Some things, just need to be developed quickly. And there's huge range of products in between.
> The issue here, as I see it, is an overdependence on electronic duck
This we can agree on. But is, at best, orthogonally related to the rest of your opinionated post.
> Do you want to look at cute icons and animations, or do
> you want to get work done?
This is a false dichotomy. You can gave cute icons and animations, and might help you get your work done faster.
The issue here, as I see it, is an overdependence on electronic duck
in preference to a Web interface to e-mail, only opting for the latter
when I have no other choice and I need to check my e-mail *now*.
I also don't like the way this guy refers to "an ugly green screen: text
only". If it gets the job done, who cares how pretty it looks? That
mainframe sure wouldn't run any faster pushing a big honking GUI out to
the terminals. Do you want to look at cute icons and animations, or do
you want to get work done?
Interoperability in the DOS-dominated era of the PC was at least more
hopeful at times, because it was not uncommon for everyone to publish
their file formats, and a software patent encumbering the file format
was just unheard of. Now, it may not even do you any good to be able to
read the file if you have to license a patent to be able to *legally* do
so. I don't see this as a step forwards.
Joel forgets to mention the main reason you couldn't copy data between
1-2-3 and Wordperfect: only one was ever running at a time. This is the
reason TSR apps like Sidekick became so popular.
Having never actually used Gmail but having tried to use Yahoo! Mail
(one of its competitors), it seems like every new version of the latter
has been increasingly more hostile towards minimalist browsing
situations. I think now you almost get laughed at if you actually need
to check your Yahoo! Mail if you only have, say, Lynx (still in active
development!). Any other time the "Web" interface to e-mail is this
screwed up, I just ssh in and use something like mutt, or even run
something like fetchmail. Yahoo! considered this a "premium" service
last I checked.
Also, there are definite security advantages from only being able to cut
Web worm could do, compared to the current generation of Windows worms.
Shawn K. Quinn <skquinn at speakeasy.net>
* * Please support the community that supports you. * *
For unsubscribe and other options, including the Tip Harvester
and archives of thelist go to: http://lists.evolt.org
Workers of the Web, evolt !
More information about the thelist