> > Phil said:
> > > we know it's being worked on. I suppose it might be better
> > > to collect url and pages of test when possible
> >
> > Well that wouldn't work for me, unless I told Olli et al
> > my Windows 2000 login password (and NO you aren't getting it!)
>
> Ok but storing informations and/or routines in a frame is
> quite common on the net. if parent is not implemented, well,
> no need to detail more no ?
Well I've suggested it should be implemented just in case..
> > > 1) might not be true on their current version
> >
> > In which case they can say "it's fixed and will be released
> > soon". It's not like they'll have to RE-fix it :)
>
> Shouldn't they finish Js before coping with bugs?
No, what development usually does is freeze a limited feature
set, debug it, release it, and then add another set of limited
features. So eventually you'll have a half-complete JS engine,
fully bugfixed, with some experimental additions. When the
additions are fully bugfixed (or no more bugs can be found :)
they get integrated into a release.
You can't just say "well it's a beta, it's SUPPOSED to have
bugs, so I'll ignore them and hope Olli fixes them for the
next release" - you HAVE to report them or they WON'T get
fixed! Take it from a programmer :)
> development time :) (I can't imagine the time it would take to Olli
> to follow a complete V development cycle at each phase for
> every line of code)
? Nobody follows the development cycle to a 't' :)
> Do you think the next pre-release will be complete and
> bugless?
No, but it will be especially bug-GED if you DON'T REPORT
THE ONES YOU FIND. Even ones that lead nowhere are useful.
> I think pre-release 2 will be even better than pre1, and for
> the rest it'll be a matter of pre-release3.
I don't get your thinking on this. Bugs don't just "disappear"
because you find them. You HAVE to report them as well :) I've
looked at the bugs database online, and it's SPARSE. None of the
bugs I've seen people moan about here (after confirming they
ARE bugs) have been reported IIRC..
> In fact I hope they won't walk towards a stable version as
> long as not everything is in, I would fear for the
> future of V.
That's just not how it works. The process of implementing
HUGE amounts of features and THEN bugfixing them is daunting
to say the least. It is MUCH easier, MUCH quicker, and DEFINITELY
more productive if you do it in baby steps. You seen the film
What About Bob?
> and that's what I'd like it to be, not an incomplete rock
> solid sequel of pre1 that will be up to date in V4 :)
As I say, that's just not how it works!
> > The JSERROR.LOG is as informative as a book with no pages for
> > the parent classes :) It just says "invalid" which obviously
> > means "not implemented yet".
>
> So..
So reporting unimplemented features and posting up 1000-line
JSERROR.LOG files to SHOW that they are unimplemented (as if
the author wouldn't know..) is stupid. So I'm not going to
do it ;)
> > > When there is a more complete version things might turn
> > > different...
> >
> > What's the point of having a beta browser with a beta
> > JS compiler if nobody reports bugs 'til the full release?
>
> Well... I'm not behind their brain actually. I just suggested it
> was no need flooding vapor with tons of bug reports at this
> stage, where the development is obviously not finished...
/me bangs head on table in frustration
if you don't report bugs, they don't get fixed. If the bugs
don't get fixed, it becomes a shitty product. It took Microsoft
5 versions of Internet Explorer to get it slightly right. I'll
let you into a secret - the reason it's so crap (actually I
quite like IE5 - I would love V3 to be better :) is because
NOBODY BOTHERED TO REPORT BUGS TO MICROSOFT.
> At this point I think remarks like 'where is this feature'
> that are more or less design related are more useful than
> reporting endless lists of js problems...
Don't be silly, now.
> I'm not saying it's no use reporting bugs, I just suggest
> some of these bug reports are too early, V3's been released
> for our pleasure and for the WOA more than for beta testing.
There is no such thing as a useless bug report. Early or
otherwise. It would be better for Olli (and I suspect he
knows this) to fix bugs first, and ADD FEATURES LATER. It
saves a LOT of work later when you find you've layered 5
or 6 functions on top of each other and the bottom
layer is full of holes - your other functions just fall
through them..
--
Matt Sealey [EMAIL PROTECTED]
Distributed Systems Support
Computer Centre
University of Leicester
____________________________________________________________
Voyager Mailing List - Info & Archive: http://www.vapor.com/
For Listserver Help: <[EMAIL PROTECTED]>, "HELP"
To Unsubscribe: <[EMAIL PROTECTED]>, "UNSUBSCRIBE"
