Category: standards

Web platform technologies in Safari 6.1 and 7

At the recent unveil­ing of OSX Mav­er­icks Apple also announced Safari 7, with great­ly improved web stan­dards sup­port. It was left a lit­tle unclear as to which ver­sions of OSX it would run on, but brows­ing through their devel­op­er area this week I found a down­load­able pre-release of Safari 6.1, which I think clears that up: it seems Safari 7 will be exclu­sive to Mav­er­icks, while 6.1 will run on Lion and Moun­tain Lion, with all of the web stan­dards sup­port of Safari 7, but only a lim­it­ed set of new features.

As both ver­sions are a major update for the brows­er, bring­ing almost a year’s worth of WebKit updates, I thought it would be use­ful to take a look through the new and updat­ed fea­tures in each, as well as try­ing to iden­ti­fy where they differ.

Read the full article

Code as she is wrote

There is a famous Por¬≠tuguese-Eng¬≠lish phrase¬≠book, pub¬≠lished in the 19th Cen¬≠tu¬≠ry, with the title ‚ÄúEng¬≠lish As She Is Spoke‚ÄĚ. It con¬≠tains many unin¬≠ten¬≠tion¬≠al¬≠ly hilar¬≠i¬≠ous trans¬≠la¬≠tions of words and expres¬≠sions, includ¬≠ing such famil¬≠iar phras¬≠es as ‚Äúthat are the dish¬≠es whose you must be and to abstain‚ÄĚ, and ‚ÄúI not make what to coug¬≠hand spit‚ÄĚ. The author, Pedro Car¬≠oli¬≠no, had the best of inten¬≠tions in pro¬≠duc¬≠ing this book, but suf¬≠fered from one major draw¬≠back: he did¬≠n‚Äôt speak Eng¬≠lish. The book was appar¬≠ent¬≠ly trans¬≠lat¬≠ed from an ear¬≠li¬≠er Por¬≠tuguese-French phrase¬≠book, using a French-Eng¬≠lish dictionary.

The rea¬≠son I bring this up is that I think this is a fair¬≠ly com¬≠mon prob¬≠lem in cod¬≠ing. Many peo¬≠ple know how to write code in order to get a result, but they don‚Äôt know the lan¬≠guage at hand in enough depth to realise that the result does¬≠n‚Äôt always make sense.

Read the full article

A Notional Proposal for a Responsive Images Syntax

As you’re prob­a­bly aware, the search is under­way for a new respon­sive images syn­tax. Matt Wilcox wrote an excel­lent arti­cle look­ing at each of the pro­pos­als and assess­ing their strengths and weak­ness­es. Read­ing this arti­cle made me con­sid­er the prob­lem, and I’ve put togeth­er a notion­al syn­tax based on the pos­i­tives and neg­a­tives of all of the pro­pos­als. I’m going to sub­mit it to the Respon­sive Images Com­mu­ni­ty Group, but first I’d like to ask you to kick the tyres a little.

Read the full article

On Opera’s Implementation of WebKit Aliases

As I‚Äôm sure you‚Äôre aware, Opera recent¬≠ly released a pre¬≠view build of their brows¬≠er Mobile Emu¬≠la¬≠tor which is notable large¬≠ly because they‚Äôve aliased a group of -webkit- pre¬≠fixed prop¬≠er¬≠ties, effec¬≠tive¬≠ly sup¬≠port¬≠ing anoth¬≠er ven¬≠dors sup¬≠pos¬≠ed¬≠ly pro¬≠pri¬≠etary code in their own.

Read the full article

An urgent call to action on vendor prefixes

On Tues¬≠day I wrote a post for on ven¬≠dor pre¬≠fix¬≠es; what they are, what they are for, their per¬≠ceived suc¬≠cess¬≠es and fail¬≠ures. This turned out to be incred¬≠i¬≠bly time¬≠ly as a few hours lat¬≠er the min¬≠utes of the lat¬≠est CSS Work¬≠ing Group were released, show¬≠ing that the mis¬≠use of ven¬≠dor pre¬≠fix¬≠es ‚ÄĒ espe¬≠cial¬≠ly -webkit-, and espe¬≠cial¬≠ly on mobile ‚ÄĒ has now become so seri¬≠ous that Microsoft, Mozil¬≠la, and Opera are all con¬≠sid¬≠er¬≠ing imple¬≠ment¬≠ing -webkit- pre¬≠fixed prop¬≠er¬≠ties in their own browsers just to ensure that their users aren‚Äôt exclud¬≠ed from the web.

What a state we‚Äôre in.

This morn¬≠ing Daniel Glaz¬≠man, chair of the CSSWG, issued an open call for urgent action by devel¬≠op¬≠ers to stop this sit¬≠u¬≠a¬≠tion from dete¬≠ri¬≠o¬≠rat¬≠ing any fur¬≠ther, and hope¬≠ful¬≠ly to improve it: Call for Action: The open web needs you *now*. I urge you to read this, and to act on it to the best of your abil¬≠i¬≠ties. If browsers sup¬≠port oth¬≠er browsers‚Äô pre¬≠fix¬≠es, the whole thing col¬≠laps¬≠es. As Daniel Glaz¬≠man says:

Ven¬≠dor pre¬≠fix¬≠es have not failed. They are a bit sub¬≠op¬≠ti¬≠mal but they also very clear¬≠ly pre¬≠served Web Authors from chaos. We can cer¬≠tain¬≠ly make ven¬≠dor pre¬≠fix¬≠es work bet¬≠ter but we can only do that if ven¬≠dor pre¬≠fix¬≠es remain VENDOR prefixes. 

Please read his post in full, and do what you can to turn this sit¬≠u¬≠a¬≠tion around. We made the mess, we need to clean it up.

An Argument In Favour Of Vendor Prefixes

Hen¬≠ri Sivo¬≠nen has writ¬≠ten a fan¬≠tas¬≠ti¬≠cal¬≠ly well-con¬≠sid¬≠ered post called Ven¬≠dor Pre¬≠fix¬≠es Are Hurt¬≠ing The Web which I urge you to read in full, as I‚Äôm about to dis¬≠cuss it. I think some of his points are absolute¬≠ly right, but I dis¬≠agree on the final con¬≠clu¬≠sion. The points that I think he nails are:

  • Demos that work in one browser/engine only and aren‚Äôt updat¬≠ed even when oth¬≠er browsers imple¬≠ment the new fea¬≠tures, are harmful.
  • Using pre¬≠fixed fea¬≠tures as a way of active¬≠ly exclud¬≠ing oth¬≠er browsers is lazy and anti-competitive.
  • Includ¬≠ing an unpre¬≠fixed prop¬≠er¬≠ty in antic¬≠i¬≠pa¬≠tion of even¬≠tu¬≠al stan¬≠dard¬≠i¬≠s¬≠a¬≠tion works against the prin¬≠ci¬≠ple of using prefixes
  • When browsers have work¬≠ing, large¬≠ly com¬≠pat¬≠i¬≠ble, imple¬≠men¬≠ta¬≠tions of prop¬≠er¬≠ties (like trans¬≠form) they should remove the pre¬≠fix even if the spec has¬≠n‚Äôt reached the stage required by the W3C.

How¬≠ev¬≠er, I still think using pre¬≠fixed prop¬≠er¬≠ties is the right approach. If we do as Hen¬≠ri sug¬≠gests and leave exper¬≠i¬≠men¬≠tal fea¬≠tures in exper¬≠i¬≠men¬≠tal builds (an emi¬≠nent¬≠ly sen¬≠si¬≠ble sug¬≠ges¬≠tion, I might add), the pace of progress will be much slow¬≠er. I believe that hav¬≠ing these fea¬≠tures out there and get¬≠ting peo¬≠ple using them encour¬≠ages com¬≠pe¬≠ti¬≠tion (and col¬≠lab¬≠o¬≠ra¬≠tion) between brows¬≠er mak¬≠ers, and the ben¬≠e¬≠fits of that com¬≠pe¬≠ti¬≠tion are giv¬≠en to us, the devel¬≠op¬≠ers, and passed on to the audi¬≠ences of the sites we build.

That does¬≠n‚Äôt mean that the cur¬≠rent sit¬≠u¬≠a¬≠tion is ide¬≠al; far from it. But improv¬≠ing things will involve more effort from us, the devel¬≠op¬≠ers and writ¬≠ers, the com¬≠mu¬≠ni¬≠ty in gen¬≠er¬≠al. Chief respon¬≠si¬≠bil¬≠i¬≠ties will be:

  • Com¬≠mit to sup¬≠port¬≠ing old demos that we have made.
  • Always make sure we plan our designs and builds in a way that they are not reliant on pre¬≠fixed fea¬≠tures, but degrade gracefully.
  • Don‚Äôt use an unpre¬≠fixed prop¬≠er¬≠ty before stan¬≠dard¬≠i¬≠s¬≠a¬≠tion has hap¬≠pened unless we can have rea¬≠son¬≠able con¬≠fi¬≠dence that it is safe.
  • Make sure we stress that the things we use and teach are exper¬≠i¬≠men¬≠tal and sub¬≠ject to change.

As an author and writer I’m guilty of some of the faults that are point­ed out in his arti­cle, and I promise to do bet­ter in the future.

In clos¬≠ing, I want to con¬≠grat¬≠u¬≠late Hen¬≠ri on writ¬≠ing this thought-pro¬≠vok¬≠ing post. I was ready to dis¬≠miss it as one of the lazy arti¬≠cles crit¬≠i¬≠cis¬≠ing pre¬≠fix¬≠es that occur reg¬≠u¬≠lar¬≠ly, but the argu¬≠ment was very well thought and well made, and made me change my mind a few times while I was writ¬≠ing this post.

Newer | Older


I‚Äôve updat¬≠ed my Speak¬≠ing page to include more con¬≠fer¬≠ences, more videos, and a lit¬≠tle on my speak¬≠ing require¬≠ments and pref¬≠er¬≠ences. I‚Äôm plan¬≠ning to cut down on the num¬≠ber of talks I give in 2014 (twelve is too many), but am always open to inter¬≠est¬≠ing offers and oppor¬≠tu¬≠ni¬≠ties, so please get in touch if you‚Äôre organ¬≠is¬≠ing an event.

[#] 1 Comment