OnlineWebCheck.com

[Albert Wiersch]

I'm in the process of upgrading http://www.OnlineWebCheck.com/ to be based on the Std/Pro editions of CSE HTML Validator instead of the free lite edition. This should provide much improved checking and CSS checking as well.

Please test it out and let me know if you encounter any bugs or problems. You can enter your site's URL or a snippet of HTML or CSS in the "Advanced Interface".

Thanks!

[MikeGale]

I really like this move. It gives the interested person a better taste.

I ran through a few pages and got good results. Nothing unexpected.

I followed the "Why Validate" link. I disagree with a statement on that page. It suggests that IE not be used. My approach is a lot different. I use a range of desktop browsers during development, mainly current ones. I develop with what my visitors are using. It's not my job to tell them what browser to use (though I may kick them elsewhere if they have JavaScript disabled). That list always seems to have IE on it.

My view is that if developers aren't seeing what IE does then they are missing something important.

(On the security issue, I think that MS may now have one of the best systems around. In sharp contrast to at least one browser plugin which I can think of!)

I recommend rephrasing that section.

(These days, for particularly important content, I also test other user agents. Like mobile devices. I do that using a service, not on my own systems. (So the tests are not as good as good as my own!) These UA's are becoming more important. It might be worth mentioning these.)

[Albert Wiersch]

I really like this move. It gives the interested person a better taste.

Great! I'm glad you like it.

I ran through a few pages and got good results. Nothing unexpected.

Also good news. Thanks.

I followed the "Why Validate" link. I disagree with a statement on that page. It suggests that IE not be used. My approach is a lot different. I use a range of desktop browsers during development, mainly current ones. I develop with what my visitors are using. It's not my job to tell them what browser to use (though I may kick them elsewhere if they have JavaScript disabled). That list always seems to have IE on it.

My view is that if developers aren't seeing what IE does then they are missing something important.

(On the security issue, I think that MS may now have one of the best systems around. In sharp contrast to at least one browser plugin which I can think of!)

I recommend rephrasing that section.

I wonder if I'm not being clear enough. I am not suggesting that developers don't test with IE. What I was trying to say was that because of IE's history of security issues, some people use and recommend using different browsers, and that means more people will be using other browsers - hence the need for more testing. IE's history of flaws is one reason (among others) that my main browser is Firefox (yes I know Firefox has its issues as well).

(BTW, I do agree that IE's security has improved.)

So knowing the above, do you have any suggestions about that part? Do you still think it should be changed, and if so, do you have any specific suggestions?

(These days, for particularly important content, I also test other user agents. Like mobile devices. I do that using a service, not on my own systems. (So the tests are not as good as good as my own!) These UA's are becoming more important. It might be worth mentioning these.)

Yes, that makes sense. I will add a section for mobile devices. Thanks for the suggestion!

[MikeGale]

I've recently found an increasing number of sites that don't work in IE. Given the large proportion of IE that I see (for some sites) this is pretty nutty on the part of those developers!

The original text reads

Internet Explorer Security Issues
Because of the history of security issues with Internet Explorer, many computer experts and companies (including AI Internet Solutions) are recommending that people not use it. This means that other browsers, like Firefox, Google Chrome, Opera, and Safari will be used more often. Therefore, your web pages should be built properly so that they can be viewed without problems on various browsers, and not just on Internet Explorer.

I suggest something more like:

Variety of Browsers
The spectrum of browsers visting a site is typically much richer than it once was. To cater to this a developer should build a site that works well in all the important ones, this might include Firefox, Google Chrome, Opera, Safari, Internet Explorer and the increasing number of mobile browsers.

I'd not try to explain how you test on the mobiles though, unless I found a reliable well curated reference.

Given that the mobile market is evolving rapidly I suspect it will get worse.

[Albert Wiersch]

I've recently found an increasing number of sites that don't work in IE. Given the large proportion of IE that I see (for some sites) this is pretty nutty on the part of those developers!

Yes, that sounds pretty nutty, but I have to admit, testing in Firefox is where I do the most testing because it's the browser I mainly use. I do test in IE, but not as much... and sometimes I might forget, especially when I think the change is minor and shouldn't cause any problems - but issues can always "sneak up on you".

I suggest something more like:

Variety of Browsers
The spectrum of browsers visting a site is typically much richer than it once was. To cater to this a developer should build a site that works well in all the important ones, this might include Firefox, Google Chrome, Opera, Safari, Internet Explorer and the increasing number of mobile browsers.

Thanks. I've changed that whole section and based it on your suggestion.

[MikeGale]

Thanks, I like that.

I have an untested suspicion that what I might be seeing (with IE8 form submit failures) is submission to some PHP sites. (Maybe their PHP is compiled without utf-8 support.) One thing I do know, they didn't test on that browser.

At an early stage of developing pages that use forms, jQuery, cookies and the like I often spend a lot of time in several browsers. Going through all the code paths, and the markup they expose. I do find things, some of them fairly odd. Some never explained!

I see interesting things like the different "feel" you get from the way Safari renders text.

Like you I often barely test minor edits.

(Side issue: I was editing Wikipedia yesterday. It was not a pleasant experience. I find their markup hard to work with on the fly. Anybody here know of easy ways to validate Wiki markup?)