Re: Navigator standard change proposal

TL;DR: You getting blamed by some people for a broken site doesn't mean  
browsers aren't getting flak for, and losing users as a result of, the  
same site. As browsers see it, the cost of losing those users is much  
higher than the benefit of telling the truth in navigator....

On Thu, 13 Feb 2014 15:04:45 +0100, Predrag Stojadinovic  
<predrags@nvidia.com> wrote:

[...some points of agreement...]
> "Their judgement is that it is easier for developers to test in multiple  
> browsers than for browsers to patch all the sites that get this wrong."  
> - I completely agree.
> I would even go as far as to say that as long as the browser does not  
> have a bug, it is not their responsibility to fix a bad site.

And here we also disagree.

If a bank site A doesn't work in browser X, but does in browser Y, then  
*many* users are likely to switch to browser Y.

When sites like Gmail or Facebook discriminate between browsers, then  
browsers will work hard and fast to make sure the site works. When the  
discrimination is not based on a real technical issue, but applied through  
browser sniffing, there are usually howls of discontent. And this happens.

Similarly, when a business chooses a piece of software for internal  
systems and it doesn't work in browser X, it typically just forces all the  
users to switch to browser y. Ironically, probably most browser developers  
have been under pressure at some time or other to use a competitor's  
product for some everyday part of their work. (Very few things annoy me  
more than this, which leads to me having a very idiosyncratic online  
behaviour. But I can afford to be more resistant than most).

> "On [the proposition that browsers shouldn't seek to stop developers
> making mistakes at the price of limiting other developers' options]
> we have a serious disagreement. It is unclear to me what sort of  
> argument would convince a software manufacturer to allow their software  
> to be used in such a way that it puts their business at risk." - see,  
> but this is where I do not understand the risk that is continuously  
> mentioned? Maybe it is because I disagree with the statement that  
> browser are or should be blamed for developers mistakes?

I agree that browsers *should*not* be blamed. Whether they are or not is  
often immaterial. The result of incompatibility is losing users, and  
losing users is a death spiral. (The smaller the market share, the  
stronger the argument that the browser doesn't matter, and the more  
compatibility problems are likely to arise. At least that's how I think  
most people working in this space interpret it…)

The alternative is increased spending on compatibility, and while that is  
the trend, above a certain level of expenditure that's an obvious death  
spiral too. Browsers (yes, including Mozilla) are businesses and have to  
work within budgets. Even augmenting the paid work with people who help  
out for free, as nearly all browsers do one way or another, there is only  
a certain amount of capacity that can be spent on compatibility.

The costs come in several ways. One is paying engineers, as most browsers  
do. Another is in managing contributions - reviewing them, soliciting  
them, identifying gaps, etc. A third is in performance. Estimates of how  
much of a browser's code is only to fix compatibility issues caused by bad  
developers range from "surprisingly high" to "the overwhelming majority".

While HTML5 has vastly improved this situation, that means it has gone  
 from awful to bad…

> From my personal experience this is anything but the case. I always get  
> blamed and more so when something does not work in users preferred  
> browser. Even when it is clearly and unavoidably the case that the  
> problem is caused by the bug in the browser or when the browser is so  
> old that it simply cannot handle the required features, it is still my  
> fault in the eyes of the user. I was unable to convince my best friend  
> to just upgrade from IE5 when already IE8 was out... he simply claimed  
> that IE5 works in all other sites he uses, so why can't I get my site to  
> work with it too...

…and I presume you know that a lot of your colleagues *did* to it too.  
Many reluctantly, but recognising that they, like browsers, have to meet  
expectations they can't force to be reasonable in order to stay in  
business, and so making hard decisions about whether to do something that  
would be really good, or to do something that will stop them losing the  
user-base that pays for them to exist in the first place.

> I just don’t see users blaming browsers... maybe that’s the problem.

Sounds quite likely. While I see users blaming browsers, and the impact  
that has on bottom lines, I certainly acknowledge that users *also* blame  
developers. And sometimes for things that are not their fault (but  
sometimes quite fairly).

Indeed, I think a success story of the Web over the last decade came about  
in part through pressure being applied to browsers and developers to  
actually use the standards correctly, through a messy and painful blame  
game that forced people to do a lot of rewriting code, and in part through  
a movement to make the standards reflect the constraint that reality is  
more important to the platform than purity of principle and elegance of  
architecture.

cheers

Chaals

-- 
Charles McCathie Nevile - Consultant (web standards) CTO Office, Yandex
       chaals@yandex-team.ru         Find more at http://yandex.com

Received on Thursday, 13 February 2014 17:16:26 UTC