Talk:Opera (web browser)/Archive 4

IE 7 = The Phantom of Opera?
Ouch. Are there more confidential news sources?
 * Update: Opera recently confirmed that Microsoft has not approached the browser maker and there is no active acquistion deal between the two companies currently - from your link. Microsoft wouldn't touch Opera. --LBMixPro&lt;Sp e ak 20:26, 23 December 2005 (UTC)
 * Opera.com would probably mention being owned by Microsoft if it were true. --Evice 00:36, 20 June 2006 (UTC)

Market share #3

 * 1) Opera is always identified -- as this shows
 * This information is very outdated. UA.ini is not even mentioned. UA.ini settings of 5 or 6 will COMPLETELY hide Opera. To overcome malicious targetting Opera has to ship with a cloak for many blocking websites by default, some of them huge ones which are used by "browser stat" sites.
 * 1) Even if someone can take extreme steps to remove that ID, then I still don't see how that can be a reason for undercounting
 * You don't think that not seeing a browser at all will lead to it being undercounted? Weird...
 * 1) There is no evidence to suggest that Opera is undercounted by the web stats sites. If you have some, provide it. Just including vague claims that the stats are wrong is not good enough. - Motor (talk) 13:39, 3 January 2006 (UTC)
 * The main reason Opera is undercounted is not that it hides itself completely, but that Opera consistently uses less connections to retreive data than MSIE and Firefox. For example, both MSIE and Firefox prefetch data: on visiting many sites these browsers will access many resources in the hope that this will speed up browsing if the user choses to visit these next sites. Especially sites that use links with the rel="next" and rel="prev" relationships will be subject to this overcounting. Jordi·✆ 12:15, 4 January 2006 (UTC)

Just in case there's any doubt -- this (above) is for discussing the questions over market share stats. I repeat: there is no reliable evidence to suggest that Opera is undercounted (note: even if you can with extreme effort cloak the UA string)... suggesting that it is without providing reliable evidence is just misinformation. If you can provide a reliable source showing that it is undercounted... fine... otherwise, your additions are not justified. - Motor (talk) 23:27, 3 January 2006 (UTC)
 * Another topic not even mentioned is that Opera actually has close to 6% market share in Europe and (parts of) Asia. The 1% is because a) most stat sites are based in the US, b) most tracked sites are US-en based, and c) Opera has very little adoption in the US compared to the ROTW. Jordi·✆ 12:15, 4 January 2006 (UTC)
 * See http://virtuelvis.com/archives/2005/05/statistics-nonsense for some more information. It is very easy to duplicate the tests given there and come to your own conclusions. Jordi·✆ 12:24, 4 January 2006 (UTC)

''You don't think that not seeing a browser at all will lead to it being undercounted? Weird...'''


 * As I made clear, and as your link showed, you have to take extreme steps to do it. The idea that this leads to significant undercounting is complete nonsense. In addition, adding claims that *web stats services* (which is where the figure we quote comes from) don't count Opera even though it is clearly identified is pretty bold -- it requires a certain level of incompetence from them... and to add it to this article you need a very reliable source.

''Another topic not even mentioned is that Opera actually has close to 6% market share in Europe and (parts of) Asia. The 1% is because a) most stat sites are based in the US, b) most tracked sites are US-en based, and c) Opera has very little adoption in the US compared to the ROTW.''


 * Had you read the talk page first, you would have seen this mentioned before. You can add in usage data from Europe/parts of Asia if you like, but it has to be from a reliable source. Handwaving hype isn't good enough.

http://virtuelvis.com/archives/2005/05/statistics-nonsense


 * Is a blog entry and pure heavily disputed opinion... no different from what you posted here (and on top of that it in no way supports the claim that Opera is undercounted compared to other browsers because of caching). We need *reliable* sources for information added to Wikipedia. What you keep adding is practically disinformation meant to suggest that the market share quoted is "probably" wrong. If it is wrong, *fine* but you need to provide realiable sources to show that this is the case -- not just assertions. If you keep adding unsourced claims, I will keep removing them. - Motor (talk) 12:43, 4 January 2006 (UTC)

Being able to completely mask the UA string with a relatively complex method that only one or two people use does not significantly alter the market share statistics of Opera. Other browsers can probably do this too, i wouldnt be suprised if there was a firefox extension for such a thing.

However, i do agree with one point made on this page: that Opera caches pages more, so less hits are made to the web servers. -- jeffthejiff  ( talk ) 14:35, 4 January 2006 (UTC)


 * But the tracking sites measure vistors (different IPs), not hits. This "caching" business is a point that has been made and debunked many times. To the best of my knowledge, Opera's caching makes no difference to how it is counted -- this is quite apart from the fact that I've never seen any real information on how Opera's cache is somehow "more efficient". It's just a claim that's thrown around whenever the subject of market share comes up. At the very least, it would take something a lot more substantial than a blog post that simply reiterates the same old stuff before I would consider including it. Indeed, this "caching " point is made in the blog post comments -- the blogger then dodges the issue by claiming to be talking about statistics in general... and that a proxy will hide lots of users making stats unreliable. A fair point, but moving away from his initial claim that Opera is undercounted... as this will undercount all browsers, not just Opera. It comes back to the simple point that there is no evidence that Opera is undercounted, and including it in the article without a reliable and respected source to back it up is simply disinformation.


 * As for the market share figures themselves. We quote Usage share of web browsers -- if anyone has problems with the figures, they should be arguing it out there really. - Motor (talk) 15:08, 4 January 2006 (UTC)
 * Oh yeah, i didnt think of it measuring the IPs rather than the hits. Oh well, I have no quarrels with the market share quotes of Opera though or whatever. They're probably not 100% accurate, but its the best indication we've got and its industry standard to use such quotes. -- jeffthejiff  ( talk ) 15:52, 4 January 2006 (UTC)


 * Will you listen for once? WITHOUT ANY TWEAKING NECESSARY OPERA COMES CONFIGURED WITH UA.INI IN PLACE FOR A NUMBER OF WEBSITES. Jordi·✆ 17:18, 4 January 2006 (UTC)


 * Which websites? You are throwing around claims that you cannot justify. I keep reverting the article back to a simple statement of market share as quoted from Usage share of web browsers. I'm not going to let you include unproven claims here. Let me make this perfectly clear: You haven't provided any evidence to justify a conclusion that Opera is undercounted -- and adding the weasel word "probably" doesn't help. The most you have done is *claim* that there are small number of websites configured in newer versions of Opera to report *WITHOUT* an Opera ID at all -- and exactly like IE/Mozilla's user agent. How do you know this has affected the stats on Usage of web browsers? You don't. The rest of the claims made in your addition were also unsourced (Europe/Asia share), have been debunked (or at the vert least are the subject of considerable controversy) such as the caching business.
 * At the moment, it's a grand total of 22 websites, based on the current contents of ua.ini (install Opera 8.51, then run help/Check for Updates and look for the file.) With a value of 4 it will pretend to be Mozilla and hide Opera completely (14 sites, mainly banks plus ebay.com, mac.com, and gettyimages.com), and a value of 5 will pretend to be MSIE and hide Opera completely (8 sites including slate.com, scandinavian.net, msn.co.jp, and a few  banks).  Values of 0, 1, 2 or 3 all include "Opera" in the identification.  While Ebay and Slate are probably high-profile, I doubt that they skew the stats as much as some people suggest. --Kelson 18:55, 17 January 2006 (UTC)
 * It's now reaching the stage where I'm simply going to advise you to take your argument to Usage of web browsers... let them argue with you over usage data. When/if you manage to convince them that Opera is undercounted, we can include the revised figures here. NOTE: If you want to include stats that show, as you claim, Opera has a much larger base in Europe/Asia -- this is fine, *if* you provide a reliable source for those figures.


 * Just because it is a blog it does not mean it is unrealible. Mozilla fanboys treat "the Asa blog as the bible, after all. Jordi·✆ 17:18, 4 January 2006 (UTC)


 * Just because it is a blog does not make it reliable enough to use as a source for your claims on Wikipedia. And what does Mozilla have to do with this? If the Firebird/Mozilla article is abusing sources in the same way you are, then I suggest you go over there and do the same thing I am doing here... making sure that unreliably (un)sourced stuff is kept out of Wikipedia. - Motor (talk) 17:42, 4 January 2006 (UTC)

Removal of the Market adoption section
If you claim the figures are misleading, take it up with Usage share of web browsers. The section quotes that article... it will also include the places where Opera is used heavily (mobile phones) and where it's technology is used in other applications. Incidentally, it would be nice to have more detail on Opera's share of the mobile market. - Motor (talk) 17:59, 4 January 2006 (UTC)


 * The problem here is that Opera has 5+ per cent market share in many European countries, so what are we supposed to believe? - Pompoms 7 January 2006 (UTC)


 * I'd quibble with the "many" part of that statement as being vague and possibly extending the Poland figures without justification. Putting that aside, I'm not sure why you see this is a problem. Global share is 0.2 to 0.9... in Poland it has managed to reach 5%. These two pieces of information are not contradictory. Which brings up another point: this assumes that the ranking.pl website quoted is reliable. We really need English language references.- Motor (talk) 22:25, 7 January 2006 (UTC)


 * It is VERY reliable. Gemius (the owner of ranking.pl) is the biggest Polish stats provider. Their stats "catch" about 10 milions of Polish web browser USERS (identified by cookies) every week from various pages. They also operate in a few other central/east european countries: Czech Republic http://www.rankings.cz/, Lithuania http://www.ranking.lt/ and Ukraine http://www.ranking.com.ua/


 * Motor, you can find some information about the company in english here: http://www.gemius.com/English/index.php
 * Some of their clients are listed here: http://www.gemius.com/English/sub.php?id=ofirmie_ref&idm=ofirmie (most of the biggest Polish portals / services etc. etc.)

Wish list - IE quick tabs
IE 7 quick tabs is damn good. It's everybody's wishlist(I love it).

Many said they will go back to IE again as long as Opera 9 doesn't provide this new Amazing feature.

Other browsers using Presto
The Wiki article lists EPOC products inclucing the Psion Series 7 and netBook as using presto but in fact they shipped with either Opera 3.62 (the first completed port for any platform to emerge from Project Magic) or Opera 5.1x (Elektra) as an extra on a CD in addition to the standard EPOC browser that was often in ROM. None of them had Small Screen Rendering.

Older Symbian communicators, Series 60 phones and UIQ mediaphones shipped with Opera 6 (Elektra), and the first releases of Opera 6 for the Nokia Communicator didn't have SSR as we know it either. There was also a cut down version of Opera 6 licensed to Motorola for a Symbian product after SSR came along that had SSR and some other features removed. It's only in the last couple of years, by which time the 'EPOC' brand had been dropped, that modern Symbian devices have shipped with Presto, be it as a free copy of Opera 7 on the accompanying CD or an unbranded version of Opera 7/8 that forms part of the device's default Internet suite (the mail client comes from Symbian, not Opera). I'm also not too sure about the Sharp Zaurus SL-5000D, which originally shipped with Opera 5. I know some later Zaurus models came with Opera 6 and some SL5000 users upgraded to that version though I'm not sure if there was an official ROM upgrade that included the new Opera or if it was unofficial versions 'borrowed' from the newer Zaurus ROMs.

I'm not sure how people would like the wiki page modified to reflect the above, if at all. This is the first time I've read the entry to be honest!


 * If you can find yourself a source to cite for the above information, you should just go ahead and change it and note where it came from. - Motor (talk) 23:28, 20 January 2006 (UTC)

About the new Criticism section
Criticism

While the point about minor updates is a valid one, I haven't seen it mentioned anywhere. The problem with pages non-compliant to W3C standards is mentioned elswhere in the article. I'm tempted to revert the change, but as a passionate Opera user I'd like to hear other people's opinions. - Discombobulatortalk 21:04, 1 February 2006 (UTC)


 * As far as I'm concerned, the section can go altogether. It adds nothing that isn't already in the compatiblity section... and the bit about updates is trivial. Quite apart from that, I dislike "Criticism" sections generally... any criticisms can be dealt with in the copy of the relevant section of the article... a criticism section just encourages people to add, well, crap. - Motor (talk) 22:50, 1 February 2006 (UTC)


 * "While the point about minor updates is a valid one, I haven't seen it mentioned anywhere." So why it could be in Criticisms of Mozilla Firefox article for several months/years and why it could be there even now? I think this critic is objective because size of updates or new verisons of Opera - (number of updates/upgrades) x ~ 4 MB against ~ 5 MB + (number of updates/upgrades) x ~ 0.5 MB (Firefox). "The problem with pages non-compliant to W3C standards is mentioned elswhere in the article." I think that this problem should be there, because it is consistent with same section in Mozilla Firefox article." If not, so there should not be in Mozilla Firefox article. Maybe this issues shloud be discussed on too. --Ptomes 16:53, 2 February 2006 (UTC)


 * We're not making this article to match Firefox, we're making it to be an article for itself. If Firefox is criticised, it doesn't have to mean Opera's criticized for it too (even if logic would dictate otherwise). Unless you provide sources, I'd call that original research. The [Firefox] article doesn't have a Competibility section, does it mean we should remove it from Opera's? The W3C problems are already perfectly well listed and explained there. - Discombobulatortalk 19:48, 2 February 2006 (UTC)


 * Criticism is fine, if it's sourced properly, valid and informative. Explaining Opera's compatiblity problems is fine, in the correct section. However, I disagree with having a criticism section by itself because it just becomes a dumping ground for people with a gripe to add one line of whining. As for comparing this article to the Firefox one... not relevant as far as I'm concerned. IMO, the Firefox article shouldn't have a separate criticism section either, let alone an entire criticism article... but since I'm not editing that article at the moment, I'll leave that upto to someone else. - Motor (talk) 20:51, 2 February 2006 (UTC)

Internet Suite
We've been over this sort of ground before... Internet Suite tells the reader nothing. Expecting them to click on a link to find out basic information is not good article writing, for a start you have to consider those who are reading printed versions too -- check the style guidelines for some examples. This opening paragraph has to tell the reader what Opera is, and what it does in a simple and clear way. If you can suggest a better opening paragraph, that's fine... but if you put this back to "Internet Suite" again, I will simply remove it... again. - Motor (talk) 16:51, 6 February 2006 (UTC)


 * I just moved the article back. Such a move should be discussed here first, obviously. --Conti|✉ 20:12, 26 March 2006 (UTC)

About deletion
I've recently deleted the section about the "Adobe CS2 and Dreamweaver MX" as well as the "Other software using presto engine"

The former section is irrelevant and is already in the Presto article. The latter section is not correct at all, some software actually use the Elektra engine instead.


 * I think the idea was to cover the whole Opera suite in this article, and if things got too big... then consider splitting it up. I'd rather see Presto (layout engine), and the awful M2 (e-mail_client) article brought into this one until that happens. As for the whingefest Criticisms of Opera, I'd like to see that axed along with Criticisms of Mozilla Firefox and Criticisms of Internet Explorer they add absolutely nothing of value to Wikipedia... they exist only provide a outlet for moans-of-the-minute from frustrated zealots and trolls. I've said this with regard to other articles too, including KDE and GNOME. Including criticisms in an article is fine (if valid)... but there is no need for a seperate Criticism section -- let alone an entire article -- as they both just encourage people to add crap (usually by including a link to some whining on a mailing list about a bug), rather than build up the article(s) by weaving relevant information into it. Usually they just end up in flamewars and feature comparisons. See Criticisms of Mozilla Firefox for a classic example. It's got a diffrent sections for zealots of each alternative browser to have a go at it. - Motor (talk) 16:40, 10 February 2006 (UTC)

Infobox 2
I've said all this before on various articles, but: IMO, Infobox 2 solves no problems, ont he contrary it causes them. It complicates editing version numbers by making it unlike editing anything else. Arguments that it avoids having to change version numbers in more than one place simply don't justify it... at best it is a sledge hammer to crack a nut. At worst, it is adds a layer of complexity and template tricks which confuse newbies (not to mention including a big "DO NOT EDIT THIS PAGE" comment as the first thing any newbie sees when editing the page)... and all for no good reason. Along the same lines, Preview versions just aren't important enough to be in an infobox -- for the same reason why I wouldn't use a preview screenshot in the infobox (unlike including preview details and discussion in the article body... where it would be welcome). The infobox is there to deliver the relevant information quickly... and preview versions just aren't relevant for most people. - Motor (talk) 16:12, 5 April 2006 (UTC)
 * Of course it solves a problem, the problem of the person editing this infobox not editing the other places the version number appears in. Perhaps you could create an Infobox 3 template that solves all problems, and we could use that. Until then, I suggest using the less problematic Infobox 2 to resolve the problem of contradictory information. -- Schapel 03:24, 6 April 2006 (UTC)


 * Why would I create yet another redundant "Infobox" when the first InfoBox already does the job? As I explained above -- Infobox 2 solves no real problems. It was originally introduced so that edits to version numbers don't show up in the page history. I have no idea why anyone thought this was a good idea, since edits are suppposed to be what page history shows. As for the matter of version numbers in different places: "Infobox 2" just creates problems by complicating editing and confusing new editors for, at best, a minor reason that could be solved with a simple HTML comment. How many other places is the exact current version number and date of Opera useful? I'd guess it's probably one place Comparison of web browsers. And because of that, the editors want to force a complicated un-Wiki system that actively discourages new editors on the editors of other articles. Wikipedia isn't a relational database and abusing templates won't make it one -- nor is it even required. I'm sorry, but Infobox 2 is just unnecessarily complicated and over the top solution to a very minor issue. That's why I removed it from the other articles I edit too and will continue to do so until someone gives a good reason why I shouldn't. - Motor (talk) 08:47, 6 April 2006 (UTC)

Original research
This edit, while well-meant is original research, I'm afraid.. So I've reverted it until a reliable source to cite can be found. - Motor (talk) 19:35, 20 April 2006 (UTC)

Extant MSN Problems?
We need a reliable source that states that MSN still sends faulty CSS to browsers IDing themselves as Opera. I know it's true (I tested it) but I can't add it because it violates WP:NOR. --M1ss1ontomars2k4 04:00, 26 April 2006 (UTC)


 * I've removed the verify tag because the article, in it's current state, is "right"... in that it is quoting a source. That doesn't mean you are wrong, just that the article as it stands is verified. See WP:VERIFY "Verifiability, not truth:


 * "Verifiability" in this context does not mean that editors are expected to verify whether, for example, the contents of a New York Times article are true. In fact, editors are strongly discouraged from conducting this kind of research, because original research may not be published in Wikipedia. Articles should contain only material that has been published by reliable sources, regardless of whether individual editors view that material as true or false. As counter-intuitive as it may seem, the threshold for inclusion in Wikipedia is verifiability, not truth."


 * Can anyone help him with a reliable reference to show that MSN is still sending faulty CSS to browsers IDing themselves as Opera so it can be updated? - Motor (talk) 19:48, 29 April 2006 (UTC)

Screenshot change?
Should we change the sshot to 9.0? I can do that, if needed. --M1ss1ontomars2k4 | T | C | @ 05:56, 17 May 2006 (UTC)
 * Opera 9.0's shot is already featured lower in the article. It should be featured as the main screenshot once it's an actual release. --LBMixPro&lt;Sp e ak 07:34, 17 May 2006 (UTC)

DOM compliance
Cannot understand the action of AdrianTM - Opera creators clame the brouser be fully DOM compatible for at least 5 years already, but it up to the last version it coud not pass the simplest tests! Users have the right to know not only pluses, but drawbacks too!


 * First of all sign your posts add ~ at the end of your post. Second, familiarize yourself with Wikipedia policies: WP:NOR -- AdrianTM 12:45, 25 July 2007 (UTC)
 * There may be a few areas of DOM support which Opera doesn't support, I don't know as that is not my field. But the forum post you mentioned dates from 2004 (and there has been no further mention of Opera there), WinLIKE development seems to have stalled and their website does not seem to be Opera friendly. If there are specific issues which relate to the CURRENT version of Opera then you need to give hard facts with proper references according to WP policies. Which is why I added a Fact tag. Dsergeant 13:50, 25 July 2007 (UTC)

1. Sorry sir, but the last post on the Opera DOM incompatibility is dated March 09, 2006 09:06AM . Since that no improvements. 2. The latest WinLIKE relise dated 06/13/2007 - is it a "stalled" soft? 3. http://www.ceiton.com/ cannot be Opera friendly by defenition - because Opera DOES NOT work correctly with WinLIKE So, I HAVE to restore my remark. If it will be deleted by you again, I'll have to call it an abuse!Stasdm 17:44, 25 July 2007 (UTC)
 * Which part of "original research" in WP:NOR don't you get? I have no freaking clue what WinLIKE is and if it follows standards, to put something like this in the article you need 1. to find a reference to it 2. the reference has to be credible, respectable, and current. 3. the issue has to be encyclopedic. Your POV pushing doesn't respect even one of these requirements. -- AdrianTM 18:13, 25 July 2007 (UTC)
 * Just as an aside, the article on WinLIKE was deleted earlier this year - see Articles_for_deletion/WinLIKE and Deletion_review/Log/2007_February_12. It is true that WinLike does not work in Opera (just checked) but the reasons are not clear, it is all buried in very unreadable javascript so may be simple incorrect browser sniffing. So WInLIKE cannot be used as an example on its own, there needs to be hard facts on the specific DOM function that Opera fails (if any). Opera 9 is largely DOM compliant, much better than earlier versions, and there has been practically no discussion about DOM on the Opera forums (where it should be discussed rather than in WP). Posts dating even from 2006 which probably relate to Opera 8 are not relevant. Dsergeant 14:10, 26 July 2007 (UTC)
 * PS - WinLIKE works in Opera if you mask as Firefox..... Dsergeant 19:02, 26 July 2007 (UTC)