A UUID function for web developers

classic Classic list List threaded Threaded
26 messages Options
12
Reply | Threaded
Open this post in threaded view
|

Re: A UUID function for web developers

denisewiseon
Actually, I haven't posted a single thing on any message board for at least 15 years, aside from two or three postings regarding some personal legal issues. I generally prefer others to do the grunt work, which I snobbishly assert they consistently do. Plus, I like shielding myself from the brutality that takes place on these public forums.

Soon after beginning web development, it was constantly hit and miss:
-No auto-grow text;
-No way to detect text overflow (what is the point of ellipses then);
-No parent selector in CSS (the parent of the element receiving the focus has the style I want to affect);
-No way to retrieve pages cross-domain (is there a point to CORS other than hindering web developers in delivering really powerful and versatile applications);
-No way to color text with a gradient brush;
-No way to have an element positioned relative to another in CSS based on its height or width at the time of rendering (top: calc(elementB.top - height));
-No nested CSS rules (.Yo:hover{a{color: red;}});
-No way to have style rules based on the element's parent client area (I want the height of one element to grow with the height of another sibling element);
-No built-in standard hashing functions;
-Getting the caret placed on an empty line of an editable box is not user friendly using a mouse;
-No style rule that can restricts what can be typed in an editable box;
-z-index has no effect on inline elements;
-No operator overloading in javascript (how can I fully take advantage of getters and setters if they don't apply to arrays);
-No "non-hacky" way to open the file save dialog (why??);
-No way to prompt the user to get the browser to take certain actions (hey, maybe I need more local storage, what is a gigabyte or two in this day and age);
-No access to the red wavy lines internally used by browsers for misspelled words.

Those are some of the browser-related shortcomings that have been beating me up in the last week. I have been encountering a streak of things I cannot do with browsers, at least not without really ugly and inefficient workarounds. I complained not once over any of them, keeping the Internet clean of my ravings, until I encountered this one unique outrage: the absence of a UUID generator function in, of all things, browsers!

This is just jaw dropping! I have an impulse to invoke the space-time continuum, odds, groupthink and the status quo, but it's all just too much; I can't separate one thought in the violent collisions of thoughts of outrage. My head reels round and round, round and round. People! People, listen to me:

I am discussing the addition of a fundamental element, something as essential as the integer.
_______________________________________________
dev-planning mailing list
[hidden email]
https://lists.mozilla.org/listinfo/dev-planning
Reply | Threaded
Open this post in threaded view
|

Re: A UUID function for web developers

Boris Zbarsky
On 3/20/16 3:11 AM, [hidden email] wrote:
> I am discussing the addition of a fundamental element, something as essential as the integer.

It pains me to say this, but the web platform doesn't really have
integers either, unless you consider typed array hacks or the sort of
things asm.js does.

-Boris

_______________________________________________
dev-planning mailing list
[hidden email]
https://lists.mozilla.org/listinfo/dev-planning
Reply | Threaded
Open this post in threaded view
|

Re: A UUID function for web developers

Wes Garland
In reply to this post by denisewiseon
>
> It is surreal I have to justify the inclusion of this, because it is
> already a basic feature in all platforms beside web browsers.


It is included in neither the C standard nor POSIX/Single-Unix
Specification.  And I daresay there is more C running out there than
JavaScript.

Also, reading this thread, it strikes me that you believe that UUIDs are
actually unique.  They are not.  They are effectively unique.

You have also not put forth even the most basic proposal, such as
identifying the TYPE of UUID you need (and justifying why everybody else
also needs this type).  Complaining and disparaging are not how things are
done around here.

Implementing ahead of standards is how we wound up in the Browser Wars,
with developers spending 500% more effort than strictly necessary on a
given task.  It has taken us basically 15 years to recover from this
state.  I'm not eager to revisit it.

Wes

On 19 March 2016 at 21:31, <[hidden email]> wrote:

> It is surreal I have to justify the inclusion of this, because it is
> already a basic feature in all platforms beside web browsers. You know, if
> just one team of web browser developers break their long-standing tradition
> of not including a UUID generator function, all others will follow.
>
> I'll consider submitting a proposal to the standards bodies, but what if
> it is rejected or takes any length of time. How can you justify the
> immeasurable number of manhours that will go to waste, because you resist
> providing a simple wrapper over a system function call?
> _______________________________________________
> dev-planning mailing list
> [hidden email]
> https://lists.mozilla.org/listinfo/dev-planning
>



--
Wesley W. Garland
Director, Product Development
PageMail, Inc.
+1 613 542 2787 x 102
_______________________________________________
dev-planning mailing list
[hidden email]
https://lists.mozilla.org/listinfo/dev-planning
Reply | Threaded
Open this post in threaded view
|

Re: A UUID function for web developers

denisewiseon
In reply to this post by denisewiseon
I agree stepping outside of standards is the last thing browsers should do, I'm not even going to target IE or older versions of browsers that lack standards support. I also admit I'm being obnoxious, forgive me for that; it's not easy pressing an issue, when the expectation was giving a little nudge and having the snowball effect keep the thing rolling by itself.

Giving the UUID a type of its own will overly complicate things. What I suggest is a function that returns a uuid in the canonical string form. Another global function can be provided to turn it into an array of bytes.
The UUID is probably the simplest construct that exists, analogous to an atom, neither divisible nor transformative. It has no business of interacting with any other type of construct. It's probably the simplest thing you can add to the platform, given its affect on other things in the platform is probably zilch. To everyone and everything the UUID is just another string, but a very unique one.

Like I've said before, this is a unique circumstance. UUID's are not only particularly suited for the Internet, but they're made for it. Browser developers have committed a grave mistake by not including a UUID generator function long ago, and there seems to be a cover up by not acknowledging this huge mistake and taking steps to correct it. Like the saying goes, "extraordinary circumstances call for extraordinary measures." This is why I insist the function should be added independent of standards.

Think of it like this: It has been virtually added over a decade ago by the community at large. Now, you'll sneak in a feature, so the community no longer has to add it manually. The community will be happy, and they'll call other browsers out for not having this very basic yet essential feature. It will become a no-brainer to other browser developers and standards bodies to include it in their work.

Whether I believe the absolute uniqueness of the UUID is beside the point; what is relevant is that I believe programmers should have access to such an entity, and allowed the privilege of believing in its uniqueness, so they can develop programs to their hearts content.
_______________________________________________
dev-planning mailing list
[hidden email]
https://lists.mozilla.org/listinfo/dev-planning
Reply | Threaded
Open this post in threaded view
|

Re: A UUID function for web developers

Eric Rescorla
I would suggest that you take a step back and ask yourself what you are
trying to
accomplish. You've now had a number of people tell you that:

- The right way to pursue this is with a standard.
- Firefox isn't likely to add this function without a standard.

You obviously don't like that answer, but I think it's a pretty clear one,
and just
repeating your arguments over and over again isn't likely to change people's
minds (though it is likely to result in them ignoring you). If what you
want is
to continue telling people they have made the wrong decision, then I suppose
that's your right, but if you actually want to get UUID added to the browser
platform, then what you're doing now is useless at best and
counterproductive
at worst.

-Ekr




On Sun, Mar 20, 2016 at 12:20 PM, <[hidden email]> wrote:

> I agree stepping outside of standards is the last thing browsers should
> do, I'm not even going to target IE or older versions of browsers that lack
> standards support. I also admit I'm being obnoxious, forgive me for that;
> it's not easy pressing an issue, when the expectation was giving a little
> nudge and having the snowball effect keep the thing rolling by itself.
>
> Giving the UUID a type of its own will overly complicate things. What I
> suggest is a function that returns a uuid in the canonical string form.
> Another global function can be provided to turn it into an array of bytes.
> The UUID is probably the simplest construct that exists, analogous to an
> atom, neither divisible nor transformative. It has no business of
> interacting with any other type of construct. It's probably the simplest
> thing you can add to the platform, given its affect on other things in the
> platform is probably zilch. To everyone and everything the UUID is just
> another string, but a very unique one.
>
> Like I've said before, this is a unique circumstance. UUID's are not only
> particularly suited for the Internet, but they're made for it. Browser
> developers have committed a grave mistake by not including a UUID generator
> function long ago, and there seems to be a cover up by not acknowledging
> this huge mistake and taking steps to correct it. Like the saying goes,
> "extraordinary circumstances call for extraordinary measures." This is why
> I insist the function should be added independent of standards.
>
> Think of it like this: It has been virtually added over a decade ago by
> the community at large. Now, you'll sneak in a feature, so the community no
> longer has to add it manually. The community will be happy, and they'll
> call other browsers out for not having this very basic yet essential
> feature. It will become a no-brainer to other browser developers and
> standards bodies to include it in their work.
>
> Whether I believe the absolute uniqueness of the UUID is beside the point;
> what is relevant is that I believe programmers should have access to such
> an entity, and allowed the privilege of believing in its uniqueness, so
> they can develop programs to their hearts content.
> _______________________________________________
> dev-planning mailing list
> [hidden email]
> https://lists.mozilla.org/listinfo/dev-planning
>
_______________________________________________
dev-planning mailing list
[hidden email]
https://lists.mozilla.org/listinfo/dev-planning
Reply | Threaded
Open this post in threaded view
|

Re: A UUID function for web developers

denisewiseon
In reply to this post by denisewiseon
> then what you're doing now is useless at best and
> counterproductive
> at worst.
>
> -Ekr
>

I recognize this is the risk I'm taking, but I've decided the best approach is to confront all the arguments presented to me. I hope everyone will jump to my side to see my perspective in the argument. I suspect the tide will turn if I can convince one mind here, other than mine's.
_______________________________________________
dev-planning mailing list
[hidden email]
https://lists.mozilla.org/listinfo/dev-planning
12