A sane presentation on future web problems

[MikeGale]

Our web has problems and I'm pretty sure they're going to continue.

Here's an interview from somebody who has some real insights into what's happening and what may well go wrong.

http://bit.ly/gB8Dgd

Doug Crockford talks about where HTML5 and other things might go wrong.

WARNING: This is a video, not text.

[Albert Wiersch]

Thanks Mike. I watched the entire video.

It made me think... should CSE HTML Validator warn users with a warning message when it detects an HTML5 document due to the issues Doug talks about (like HTML5 being "1/2 baked" and not finished)?

I think I will also add a new tip message about HTML 5's issues when someone uses the HTML 5 template.

I thought it was also intereting that he said IE9 could be the best browser in the world, like IE 6 was when it came out. I still have not experimented with it yet.

[MikeGale]

There is so little knowledgeable and sane debate on this issue. I was delighted to find it.

It made me sad though. I have reservations about HTML5, he confirms some of my fears, and expands on them. I'm impressed just how much he knows, so much more than I do!

I like to live in hope but I see a potential train smash coming along.

I have tested IE9. Kept my opinions to myself up till now. Yes I think it's potentially great. If we're lucky we'll have a return to the days that led up to IE6. Hot technological innovation, but hopefully without the incompatibilities.

I think carefully phrased warnings about HTML 5 are a good idea. (With luck the warnings can fairly rapidly (2 or 3 releases) fade away.) Of these the issue of video codecs is especially a problem and will make the tag worthless, unless the commercial interests of some companies are held in check.

I remember that Hixie said, some time ago, the specification would be fully baked in about 2022. Hope that premature popularity doesn't mess our best hope for real progress in a decade.

[Albert Wiersch]

I think carefully phrased warnings about HTML 5 are a good idea. (With luck the warnings can fairly rapidly (2 or 3 releases) fade away.)

Thanks Mike. I've added a message like this:
As of this writing (December 2010), HTML5 is still in development and may be significantly changed (it may take years before HTML5 is finished). What works now may not work in the future. Consider using HTML 4.01 Strict instead, as it is a mature specification. If HTML5 is used, then be sure to understand the potential problems with using standards and features that are still in development and that may be significantly changed.

Of these the issue of video codecs is especially a problem and will make the tag worthless, unless the commercial interests of some companies are held in check.

I hope that they are... software patents seem to cause more trouble than they're worth, at least that is my impression.

I remember that Hixie said, some time ago, the specification would be fully baked in about 2022. Hope that premature popularity doesn't mess our best hope for real progress in a decade.

Only 12 more years. Maybe in two or three decades HTML6 will be finished.

[MikeGale]

I hope that they are... software patents seem to cause more trouble than they're worth, at least that is my impression.

There are some major problems with the patent system. They're bad enough to threaten civilisation I think. A lot of Software patents wouldn't have been granted if the patent clerks had any idea of what they were doing. Worse still is being able to patent genetic material. This is so incomprehensibly stupid I don't know where to start!

My understanding is that the lords of the patent world are not really answerable to anybody. I don't know how to fix that. Politicians aren't the answer!

[MikeGale]

I saw an article about Microsofts HTML 5 labs, yesterday.

Today I saw this article http://bit.ly/hmBQ1G which puts it all in a nutshell, very nicely.

In essence it's a way to test the flakier bits of HTML 5, in a Microsoft Browser. The general public doesn't get those bits. The dedicated developer can get them if he wants. A very good idea. This sort of thing will reduce the worst dangers from a standard that is not yet ready.

[Albert Wiersch]

There are some major problems with the patent system.

I think I'll have to agree 110%! But don't get me started on that.

[Albert Wiersch]

Today I saw this article http://bit.ly/hmBQ1G which puts it all in a nutshell, very nicely.

In essence it's a way to test the flakier bits of HTML 5, in a Microsoft Browser. The general public doesn't get those bits. The dedicated developer can get them if he wants. A very good idea. This sort of thing will reduce the worst dangers from a standard that is not yet ready.

I was just able to take a look at the link you provided. I agree that it's a good idea to keep "half-baked" HTML5 stuff out of "official" public-use browsers so that developers don't start using it only to see it change or cause problems, while including the "stable" HTML5 features that can more safely be used.

Glad to see that IE9 is doing this.

[migman]

Our web has problems and I'm pretty sure they're going to continue.
Here's an interview from somebody who has some real insights into what's happening and what may well go wrong.
http://bit.ly/gB8Dgd
Doug Crockford talks about where HTML5 and other things might go wrong.
WARNING: This is a video, not text.

Thanks for sharing that, very useful information!