3 ms·
It's certainly discussion worthy though and I think discussion is more often the point here on HN than the actual "news" articles. In my opinion, the web stand
by wayneftw 8y ago
It's certainly discussion worthy though and I think discussion is more often the point here on HN than the actual "news" articles.
In my opinion, the web standards have moved way too slowly and that is directly the fault of Mozilla, who won't let us have things like WebSQL due to their ridiculous insistence on having 2 independent implementations (as if we don't already have that for SQL and, as if there has to be 2 implementations of oxygen for air to be breathable...)
I don't trust Mozilla to do the right thing here or ever anyway. They're allegedly a group that fights for merit, freedom and openness but they consistently work against those things by rejecting the best ideas and by selling their users out to marketers. Honestly, I trust them less than Google or Microsoft and since I don't like anything about Firefox anyway, I simply ignore it and build all my apps to target Chrome. And that's the same thing that a lot of people are doing, which is why web devs keep using things like WebSQL in their projects.
- Yoric 8y ago> In my opinion, the web standards have moved way too slowly and that is directly the fault of Mozilla, who won't let us have things like WebSQL due to their ridiculous insistence on having 2 independent implementations (as if we don't already have that for SQL and, as if there has to be 2 implementations of oxygen for air to be breathable...) That's an interesting view on the problem. If you recall, WebSQL not only didn't have two independent implementations, but it also didn't have specs (if you want to check, you can read the proposal here https://dev.w3.org/html5/webdatabase/ https://dev.w3.org/html5/webdatabase/ and realize that it's missing all the important stuff such as what queries do). WebSQL was a nice hack/proof of concept by the Google team, but it was formulated in such a way that pretty much guaranteed that you would eventually lose data and/or that your programs would break. That's the reason for which specs are needed, rather than just an implementation. Having two independent implementations is the only way known to humanity to ensure that the specs are useful. > I don't trust Mozilla to do the right thing here or ever anyway. They're allegedly a group that fights for merit, freedom and openness but they consistently work against those things by rejecting the best ideas and by selling their users out to marketers. Honestly, I trust them less than Google or Microsoft That's a shame. I don't think that anybody claims that Mozilla has always made the right choices – starting with Mozilla itself. However, Mozilla has consistently attempted to fight for the users of the internet. I don't remember the word "merit" ever being mentioned, though. While Google (and recently Microsoft) have better PR teams (and lots of money to pay for them), their track record is not even close to being as good as Mozilla's. > and since I don't like anything about Firefox anyway, I simply ignore it and build all my apps to target Chrome. This is certainly your right. But this is also a way to ensure that Google can make all decisions for the web, without having to go through standards. You may have already witnessed APIs that work only for Google properties. I may be wrong, but I suspect that it's only going to get worse.