He starts out by saying: “Fair warning. You’re going to hate this one.” Well, not me. I couldn’t agree more with his sentiments here. I’d like to quote about 15 different things from the article, but I’ll just do a few:
“Why should we push the web forward? And forward to what, exactly? Do we want the web to be at whatever we push it forward to? You never hear those questions.”
“[W]e need the time for a fundamental conversation about where we want to push the web forward to. A year-long moratorium on new features would buy us that time.”
“We’re pushing the web forward to emulate native more and more, but we can’t out-native native. We are weighed down by the millstone of an ever-expanding set of tools that polyfill everything we don’t understand — and that’s most of a browser’s features nowadays.”
Lots of sound reasoning in this one and some pretty harsh words for the development team working on Chrome! Regardless, I think this is a must-read for everyone in the industry.
So what do you think? Is PPK’s advice here sound? Is his described “moratorium” a solution? Or is he discussing a non-existent problem?
Not all of the changes ARE pushing the web forward. For example HTML 5 has reintroduced a lot of HTML 3.2 attributes that were marked as obsolete back in 1997 and where support for the proposed replacements is now finally available across all current browsers (including IE8). A backwards step just after it finally became possible to move forward.
Just because it took about 15 years for browsers to catch up and fully support HTML 4 was no reason for reintroducing many of the obsolete ways of doing things just after the new standard way became usable.
Didn’t Guido Rossum put a moratorium on Python a couple of years ago to kind of figure out what direction he really wanted it to go. Did it really do the Python world that much good by doing that? I see similarities, but maybe at a much more complex level.
You mean the <blink> element that put a lot of epileptics into hospital? Are you sure you can afford to pay for all those hospital charges and compensation for the harm caused?
But marquees haven’t ben discontinued - just moved to JavaScript where it belongs - plus it works there in all browsers and not just in IE.
For example the marquees at the top and bottom of http://www.felgall.com/jstip109.htm simply have class=“marquee” added to the appropriate tag and a script added to the bottom of the page - both marquees are running from a single script and I could have added class=“marquee” to more elements in the page to make more marquees.
That might not be the one I was thinking of - perhaps it was a bad dream. I seem to recall an element that allowed the border to flash, or be highlighted in kind of carney flashing border lights. Might have been just a popular ‘design’ convention at the time…
I wasn’t really convinced by the article. Some of theresponses were more convincing.
I don’t like the speed of change, but it seems worth experimenting and seeing what’s possible. In the long run, some things will take off and other things won’t. Telling people to take a break is a bit like me wishing my puppy would stop growing for a while, so that I could enjoy her for longer as she is. Nature doesn’t work like that, of course.
I share the view about the over-use of polyfills and tools: it’s getting a bit out of control. I remember the days when you only needed a CSS and perhaps a JS document to attach to your HTML and you were good to go. Now I can’t even count the number of dependencies anymore, we even need new tools to manage these.
A moratorium? I’m not sure, you can never be sure how things are going to pan out. I remember the times when you had no way to add rounded corners and shadows without complicated placements of images and extra markup. In those days, it seemed most designs needed this stuff. Then, CSS3 allowed us to do this in a split second and as soon as most browser caught up with the technology the trend changed: flat design came up and rounded corners and box shadows went out of the window.
At times I think this field needs movement, it could be movement towards something definite but perhaps it’s not, it’s movement for movement’s sake, it’s in its nature.
I feel like this most closely sums up my attitude. It’s natural progress. As long as we can do our best to discard the bad ideas as quickly as possible and continue forward with good ones… that’s the goal I guess.
Change is always inevitable, that’s a given, but the current round feels more like being in a marathon than ever before. Normally you have a few front runners, a mass in the middle and a few stragglers… this time round it feels like we have a few front runners, a few in the middle and a mass bringing up the rear!
The end result is a proliferation of CMSs, tools and templates which address the shortfalls in knowledge. Faster broadband speed hasn’t helped, hiding the resultant ‘bloat’ that’s going on behind the scenes, allowing humans to do what they do best, cut corners!
On the other hand, we cannot and shouldn’t halt progress, and the likes of Chrome breath life into the Versioning and SitePoint email links (yes I have cut out all the others) that I follow and often marvel at. So hats off to Google for that, but you are only as strong as your weakest link, so pushing for the adoption of new standards before the market is ready is counter productive, and I don’t think it can all be blamed on the lack of ‘cross browser support’!
Is Google’s advancement of web standards a genuine passion for the industry or a commercial decision to crush its competitors…? Then again, should a commercial company have a responsibility to an industry or to its shareholders?
I think education is fundamental - giving all those ‘stragglers’ (me included in some case!) time to catch up and really hone their skills with the latest ‘tools of the trade’. It’s why I joined SitePoint. But will it happen? I suspect not in a voluntary market without regulation or recognised qualifications and on-going learning support.
I found a lot in that article to agree with. I have, for a very long time, been frustrated and annoyed by developers who do things “because they can”, and who put fancy bits and pieces into their products (whether it’s a website or software, it’s the same outcome) that use the latest technological advances … but with no rationale or benefit to it, and often a significant cost. Usually it’s just a developer showing off a new trick that he has learned … and you know what? No-one else cares.
With native apps and programs, the cost usually comes in a significant performance lag for anyone with a lower spec machine than the developer – and since developers usually get the best computers going, that means pretty much everyone else. With websites, performance can still remain an issue, you also increase the risk of incompatibility, with users’ browsers not being as up-to-date as the developer’s, or maybe being just as up-to-date but implementing code differently.
When there’s a real, tangible benefit to the new feature then yes, by all means lick the edge of the envelope (or whatever the stupid phrase is) and explore the limits of what’s possible. But if you’re just doing it to show off a new coding trick that you’ve picked up, or to make your product look a little bit spiffier, have a serious think about what the implications for your users are going to be.
The Internet does “touch” more and more people all the time.
And it does seem the attitude has shifted from “I’m reinventing to learn” to “I’m reinventing because, just maybe, …”
Hence an over-whelming number of advances and a technical landscape that it is becoming more difficult to keep sound footing on. Specializing helps, just the same, at least for me, “keeping up” sure isn’t easy.
So this moratorium is a cry to “give us humans time to absorb and evaluate?”
That appeals to me. but I highly doubt it will happen.
The thing I love about the web is that you don’t have to go along with the crowd. So people are falling over themselves dealing with the proliferation of new technologies? Good luck to them! I’m happy tinkering with HTML, CSS and little bits on JS and PHP, with my little code editor and FTP client, pushing stuff to my lil’ ol’-fashioned server. Suits me just fine and is a perfectly valid way to use the web … and as far as I can see, will be into the future. So people are welcome to burn themselves out trying all the new stuff if they like. Later on, I’ll just come along behind them and use the few things that are actually useful, when they are well enough supported.
Yes, browsers still support HTML 3.2 because that’s what 95%+ of web sites still use. They have added support for things like XHTML5 and CSS3 because there are people who want to use those too.
Well we are still waiting for IE8 to die so that we can start using XHTML5 even though it has been out for almost a year (nearly 16 years of you count XHTML 1 as well).
Also most of CSS3 is yet to become a standard - only some parts of it have been finished.
Compared to the HTML 3.2 that most web authors use both of those are very new. That there are plenty of even newer things around (such as ES2017 - the version of JavaScript currently under development) just shows how web tools are shifting away from the 1997 version that most web authors appear to prefer.
@una posted this article on Twitter earlier today. It takes a slightly different slant to its take on things, but raises similar concerns regarding the web’s direction. It makes for an engaging read, even if the talk it was culled from is 12-months or so old.
