The Active Network
ActiveMac Anonymous | Create a User | Reviews | News | Forums | Advertise  
 

  *  

  Opera cries foul against MSN--again
Time: 19:14 EST/00:14 GMT | News Source: CNET | Posted By: Bill Roach

Rendering glitches in Microsoft's flagship MSN.com site have sparked renewed criticism from a competitor that the software giant is undermining its browser. Opera Software on Wednesday said Microsoft has been sending its browser a faulty style sheet, which determines the presentation of graphics and text in a browser window. When people using Opera 7 browser software visit MSN.com, published by Microsoft, some of the site content is obscured, Opera Chief Technology Officer Hakon Lie wrote in a posting to the company's Web site.

Write Comment
Return to News

  Displaying 1 through 25 of 354
Last | Next
  The time now is 11:47:53 PM ET.
Any comment problems? E-mail us
#1 By 2332 (65.221.182.3) at 2/5/2003 8:48:59 PM
Ya, after all... it's MSN's responsibility to be sure that their site renders on Opera, which has a massive .02% market share. While their at it, they better do the same for Konqueror... with their .01% market share.

Give me a break. It's barely arguable they should support Netscape/Mozilla, much less these fringe browsers.

This post was edited by RMD on Wednesday, February 05, 2003 at 20:49.

#2 By 3339 (64.175.42.78) at 2/5/2003 10:06:57 PM
More importantly, how 'bout this quote: "When asked why Opera was sent a different style sheet, when the IE style sheet would render properly in Opera, Visse said only that the Opera style sheet was the same one that MSN sent to other, earlier versions of IE."

In other words, the IE style sheet works fine, but MS specifically chooses to send Opera users a bad style sheet. Their defense is--it's the same as the old one, but that means they must know it is less feature rich. They also say they send different style sheets to provide the richest experience... well, one would imagine the way they do this is develop the IE one first, test it on other browsers, and modify it accordingly. If they had done so, they would have known that the IE style sheet works fine. If they did build a special style sheet to provide a "good experience" for non-IE browsers, why would it be the exact same one as the older version?

#3 By 3339 (64.175.42.78) at 2/5/2003 10:14:46 PM
What a surprise, macross and Jagged chasing my tail with nothing better to say.

#4 By 3339 (64.175.42.78) at 2/5/2003 10:40:45 PM
I'm curious abou the theory, diaphanein, but I do not see how web controls and .Net specific elements would be defined in a style sheet. Nor do I see why, even if there is a detection difference, they couldn't send the same style sheet (if it is accurate that the IE style sheet renders properly on IE). Correct?

#5 By 2332 (65.221.182.3) at 2/5/2003 11:00:02 PM
Why is this even an issue? Opera has about 7 people using it.

If Opera can handle everything IE can handle, it should just spoof the user agent string like Mozilla does.

#6 By 3339 (64.175.42.78) at 2/5/2003 11:48:15 PM
0RMD, please read: http://people.opera.com/howcome/2003/2/msn/

and there goes that theory, diaphanein...

Opera is spoofin the same userstring as IE (Mozilla by the way). Opera identifies further information about itself for natural reasons--that is, the user agent has a specific purpose and there is a standard for its use; you can add additional information as well so that you can actually determine if your product is being used, as one example.

Let's look at that: for one, this is simply not the standard method. Secondly, this denies Opera from testing the stylesheet themselves, determining if the IE one works, and saying, "We work like IE wants us two," and "we also want to statistically track our product."

Also, Hakon has examined the stylesheets, and the only thing that is off is shifting a margin 30 pixels in the wrong direction.

Relatively minor, right? But why is it happening at all? MS is particularly looking for Opera and sending them a stylesheet that doesn't work.

Next, Hakon has determined that the IE6 stylesheet works perfectly fine for Opera. Additionally, he's shown that if IE can't identify the browser, it defaults to the IE6 stylesheet. This sheet works, but Opera cannot get this default sheet or the IE6 sheet.

This tells me Opera is specifically being fcked with. If they weren't, MS should simply treat it as unidentified. If they are trying to treat it special (and they are--they are looking for it), that special treatment should have another purpose besides displaying it incorrectly. Particularly when it would display correctly, if it wasn't treated uniquely. MS is doing extra work to make sure things do not display correctly in Opera.

Marketshare shouldn't be an issue--that's the most pathetic excuse, RMD. (Come on, seriously... Are you saying that if there were alot of Opera uses, hypothetically, even, let's say, a majority of users, it would be wrong or something? You're rational enough to understand that whether or not MS is doing something wrong, the right-or-wrong-ness is not related to size.) If anything, the fact that they are targeting such a small segment of the marketplace, it shows how pathetic, paranoid, and underhanded they are being. They are specifically looking for Opera and mis-displaying it. Opera has to work to have their browser work like IE, and then MS does more work to make sure that Opera is treated as if it behaves differently than IE which it doesn't.

This post was edited by sodajerk on Wednesday, February 05, 2003 at 23:54.

#7 By 2332 (65.221.182.3) at 2/6/2003 12:50:51 AM
#11 - "Marketshare shouldn't be an issue--that's the most pathetic excuse, RMD."

Well, my assumption was that Microsoft wasn't doing this to Opera on purpose. My argument about market share is perfectly valid if it is simply a case of designing the site to appear best in the target browsers. (Target browsers typically being those with the most market share.)

If Microsoft is specifically trying to sabotage their site when it's viewed with Opera, then that's a different matter.

But I'm not going to just accept what Opera says. People have claimed many times that Microsoft has sabotaged their software in one way or another, and those claims have turned out to be either mistakes in programming (like Real software), or simply false. Opera would get a lot of press by claiming that Micrsoft is doing them wrong...

At any rate, I'll defer judgment on the subject until I see some independant tests.

One has to wonder, however, why Microsoft would do this. It seems to me that there are plenty of other browsers that are far more of a threat (Mozilla/Netscape), and targeting a browser than so few people use while at the same time risking press just like this seems to me to be a really stupid idea.

#8 By 135 (208.50.206.187) at 2/6/2003 1:02:42 AM
Leave it up to sodajerk to take any opportunity he can, no matter how trivial, to bash Microsoft.

I bank at Wells Fargo, when IE 6 was first released their online banking website would not even allow me to connect. I had to wait about a week for them to "certify" IE6 worked with their website.

Obviously this was all some massive conspiracy by Wells Fargo to disadvantage those of us who immediately downloaded IE6. There couldn't possibly have been any reasonable explanation for this action, oh no!

#9 By 3339 (64.175.42.78) at 2/6/2003 1:10:20 AM
RMD, it's happened before. This is the second time, and it's come out because each time Opera has clearly demonstrated exactly what's going on. I doubt Hakon, the creator of CSS and important member of W3C, wants to fck around with his reputation by being wrong.

I also bet this'll work out fine. Hakon has a clear argument and proof. MS will fix it, say its a mistake, etc. Yes, Slashdot and some other communes will be in a tizzy for a day or so, but Opera usually isn't too petty. Hell, yeah, they are documenting and looking into this stuff, but I doubt Opera would try to sue MS, see... (stu, yes, their pathetic-ness surprised me--what they said about Apple, and trying to get them to license Opera for Safari--laughable--but the money Opera does get--it gets from the embedded markets mostly, a blip of MS users, etc... I can't see them disappearing just yet...)

But I digress... On the other hand, if they fcked with Moz, you get a lot of idiot attacks and responses, and AOL will get involved, etc... Must better to test the waters on Opera. (Remember the first time a while back, it was Opera and some version of Mozilla.)

#10 By 135 (208.50.206.187) at 2/6/2003 1:28:53 AM
So I downloaded Opera... hey, it works reasonably well. The MSN homepage isn't formatted quite correctly, most of the data is there but Opera is rendering it incorrectly.

The user interface pretty much blows chunks, but once you figure out you have to do a File->New Page in order to type in a web address it's easy to go from there.(Is that unintuitive or what?)

Also had to turn off the "hotlist"(aka favorites) bar on the left by going View->Hotlist and then Off on a sub-sub-menu. Apparently they haven't figured out that it's accepted practice to be able to close such menus with a little X up in the corner.

Oh, and then the bookmarks have been conveniently filled in with like 200 some websites, have to go in and delete that. Sheesh I thought IE was annoying with it's 12 media favorites or whatever.

Also fullscreen doesn't work on my secondary monitor.

It's got some nifty little features. The throwback to MDI style browsing is ok. It can't save a snapshot of the webpage all in one file like IE, which is disappointing but it at least does render pages as fast as IE.

Ohwell, good to have some competition, but it still needs some polish.

#11 By 61 (65.32.170.1) at 2/6/2003 1:33:10 AM
Or it could be due to the fact that Opera 7 is brand new and has a bug in it.... but I guess they don't want to admit that.

#12 By 2459 (24.170.151.19) at 2/6/2003 1:42:07 AM
Why does Opera use a redundant user agent string?

Why does it bother to list itself as both MSIE5 and MSIE6 in the same user agent string?

#13 By 2459 (24.170.151.19) at 2/6/2003 1:55:43 AM
Also, is one of the shots on that site Opera 6? There are two screenshots showing Opera going to the same msie6.html page, yet one renders the page correctly, while the other is missing graphical elements from the Sign In button and the MSN8 ad.

#14 By 61 (65.32.170.1) at 2/6/2003 1:59:50 AM
Wow, I'm not the only one up this late....

#15 By 3339 (64.175.42.78) at 2/6/2003 3:50:19 AM
It's not late yet.

#16 By 2332 (65.221.182.3) at 2/6/2003 9:09:49 AM
#14 - "RMD, it's happened before."

It happened before with Opera? It happened before that Microsoft specifically targeted a browser so that it wouldn't work with MSN, and this was proved?

Have a link?

#17 By 2459 (24.170.151.7) at 2/6/2003 11:07:24 AM
It doesn't pass IE's exactly.

This is Opera 7.0's user agent string:
Mozilla/4.0 (compatible; MSIE 6.0; MSIE 5.5; Windows NT 5.1) Opera 7.0 [en]

This is IE's:
Mozilla/4.0 (compatible; MSIE 6.0; Windows NT 5.1)

(If .NET were installed, it's runtime version would also be appended)
Mozilla/4.0 (compatible; MSIE 6.0; Windows NT 5.1; .NET CLR 1.0.3705)

Netscape's:
Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.0.2) Gecko/20021120 Netscape/7.01


This post was edited by n4cer on Thursday, February 06, 2003 at 17:26.

#18 By 3339 (65.198.47.10) at 2/6/2003 1:27:40 PM
Actually enforcer, they are the same. That is: Mozilla/4.0 is the user agent header. What appears in paranthesis is the extended user agent header. This is more or less a comment field as all of them are displaying browser compatibility info rather than the intended client-side user agent properties (Netscape is mixing--I think some of the client-side props are CPU, media type, input device, resolution, language, etc.--and are specifically intended for non-computer devices) There's really no reason for MS to be using the whole user agent string--particularly the elements outside the comment (in parathensis) which are becoming a more common way of tracking browser usage.

#19 By 6253 (12.237.219.240) at 2/6/2003 2:20:12 PM
I can't believe so few people seem to recognize that this is a browscap issue. Opera probably knows it, too, but they also know that MSN is bureaucratic enough and the press is dumb enough that Opera can characterize the situation as being non-technical.

You people with IIS5 boxes, go look at your \WINNT\system32\inetsrv\browscap.ini files. Betcha it's from 1999. Now compare with the one from http://www.garykeith.com/data/browscap.ini. A little bit missing, huh? I guess you're not keeping your browscap.ini file updated along with security patches, media players, Nero updates, and all the other crap you need to keep up with.

Luckily, you're probably not running a web site which uses browscap. If you were, you'd be screwing every browser on the planet including Microsoft's. Perhaps Microsoft's IE group should sue the IIS group because the current W2K media still ships with a 1999 browscap.ini.

In MSN's case, they're not quite that out of date, but it's pretty misleading to say that they're screwing Opera on purpose and that it's not a technical issue. If you understand how browscap does its matching, you can easily see how this problem might occur if MSN isn't keeping up with all the alternative browsers out there.

#20 By 2459 (24.170.151.19) at 2/6/2003 5:22:00 PM
SJ, if they didn't check the part outside the parens, they wouldn't be able to adjust the content for the browser. The only other way they'd be able to tell Opera is Opera from it's user agent string would be to look for the redundant MSIE version listing.

If they didn't check the info inside the parens, they wouldn't even be able to target IE.

Plus, as you say, the info after the parens can be used to track browser usage. Maybe MS (or anyone else that checks the full string) is also intrested in tracking viewer statistics.

#21 By 3339 (65.198.47.10) at 2/6/2003 5:23:17 PM
holedup, that makes no sense, as far as I understand it. If a user agent string isn't identified in browscap.ini, the server should treat it as a default--special behaviors are only passed if is overtly noted. Since Opera is being treated differntly, it can't possibly be because it is defaulting or using an out of date .ini because this is a new string. Since this is a new string, and it is being treated differently, it's not a matter of just being out of date. MS updated how it is treating Opera.

#22 By 3339 (65.198.47.10) at 2/6/2003 5:32:34 PM
enforcer, I don't understand this logic. Opera is built to comply with being identified as Mozilla. Inside the parathensis, they are stating what code should display properly if it was designed for that code. Outside they are just saying--even though we comply with Moz's layout and are compatible with code developed for these browsers, we are Opera.

"they wouldn't be able to adjust the content for the browser." They ARE doing SO, and DISPLAYING it WRONG. If they just looked as Moz, it would display FINE. If they looked at the compatibility and looked for IE6 and used the IE6 stylesheet, it would display CORRECTLY. If they used the DEFAULT, it would DISPLAY CORRECTLY.

"The only other way they'd be able to tell Opera is Opera from it's user agent string would be to look for the redundant MSIE version listing." The point is they DON'T need to tell it apart; Opera is DESIGNED to display correctly even if the site is coded for IE. All you are doing is supporting my argument: MS is specifically looking for Opera and displaying it incorrectly, even though they don't need to do this extra work and Opera "certifies" itself as displaying IE6 code correctly.

"Maybe MS (or anyone else that checks the full string) is also intrested in tracking viewer statistics." This is it's purpose, yes. What does that have to do with displaying an incorrect stylesheet? Nothing. This isn't a story about browser usage stats being incorrect because everyone pretends they are IE; it's a story about MS directing Opera browsers to a different and faulty stylesheet even though the default stylesheet would be correct for Opera usage.

#23 By 2459 (24.170.151.19) at 2/6/2003 6:14:46 PM
SJ, despite the current example with the style sheets, the reality is that different browsers have differet featuresets and different levels of support for those features.

Pages frequently use VB or ECMAScript, or other code that will not work correctly on all browsers. Rather than leave non-IE or (IE on other platforms/older versions) users out due to incompatabilities, MS tries to make their site experience similar for as many users as possible within the limitations of the browser used. To do this, they detect the more popular browsers and push the content in a form fitting for that browser.

Though Opera 7 may be able to handle the page, has anyone checked to see what sheet older versions of Opera (particularly 5/6) are pushed, and whether the sheet displays correctly in those versions, and if the ie6/mozilla sheets also display correctly?
Though Opera (whatever version) may identify itself as Mozilla compatible, users have had rendering errors due to bugs, etc., that detract from its compatability. To work around these errors, or just limitations, some sites specifically target the browser.

MS is doing no different than many other sites in looking at the full string for compatability. It just happens that there was a mistake in the style sheet (assuming that sheet wasn't pushed to other versions of Opera that may have had problems that required adjusting the text position).

http://www.albin.net/CSS/OwenHack.html

#24 By 3339 (65.198.47.10) at 2/6/2003 6:38:04 PM
Enforcer, this is just silly: "despite the current example with the style sheets, the reality is that different browsers have differet featuresets and different levels of support for those features." Okay, but THIS EXAMPLE IS ABOUT STYLESHEETS, and your comment is completely irrelevent and unrelated. What's your point? That you don't have an argument so you are just blathering?

"Pages frequently use VB or ECMAScript, or other code that will not work correctly on all browsers. Rather than leave non-IE or (IE on other platforms/older versions) users out due to incompatabilities, MS tries to make their site experience similar for as many users as possible within the limitations of the browser used. To do this, they detect the more popular browsers and push the content in a form fitting for that browser." Much more irrelevent crap so please don't be offended by the word "blathering"--you are.

"Though Opera 7 may be able to handle the page, has anyone checked to see what sheet older versions of Opera (particularly 5/6) are pushed, and whether the sheet displays correctly in those versions, and if the ie6/mozilla sheets also display correctly?" Yes, yes, yes.

"Though Opera (whatever version) may identify itself as Mozilla compatible, users have had rendering errors due to bugs, etc., that detract from its compatability." Again, completely irrelevent--it is easy to demosntrate that there are NO, ZIP, ZERO rendering errors. You don't have to keep repeating obvious and irrelevent crap.

"To work around these errors, or just limitations, some sites specifically target the browser." Yes, they actually "WORK AROUND THESE ERRORS"--in this case, MS is working around the correct code to PRODUCE ERRORS.

"It just happens that there was a mistake in the style sheet" yes, i"it just so happens..." And MS has been getting called by journalists all day and have needed to respond to the issue. Bob Visse has been providing quotes and saying smarmy things like: "We have different style sheets and different code for various browsers... That's something we do to try optimize the experience for our users." Yeah, that's great Bob--I could change the pixel offset to zero and click save faster than you can repeat that sentence--why the fck hasn't it been fixed yet if you are OPTIMIZING the EXPERIENCE for me.

#25 By 3339 (65.198.47.10) at 2/6/2003 7:55:30 PM
Aaaayyy! It's fixed. I wonder if Bob read my post.

Write Comment
Return to News
  Displaying 1 through 25 of 354
Last | Next
  The time now is 11:47:53 PM ET.
Any comment problems? E-mail us
User name and password:

 

  *  
  *   *