Editable USER AGENT

Editable USER AGENT

be able to spoof like google-bot

like in firefox should be fine...

+1

+1

+1

What would be the use of this?

It is editable, google it

+1

I could not use google_bot as UA ...

how to do it ?

Even if you try and identity yourself as GoogleBot, you get nothing of the advantages unless connecting to dumb servers. Google bots -- as well as other crawler systems -- only have a well specified range of IP addresses from which they can operate an access remote content. No matter what kind of connection to the internet you have, chances are 100%, ie.:your provider won't let you, that your address won't match the specified range. Thus you'll get nothing, just be a stranger whom server admins look and say, "oh look at that n00b! He thinks he's google and can access out content!"

General ID spoofing (as, say, updating the current user agent string options) would be a good thing to have. But it can't help you masquerade as a crawler.

Originally posted by RyanChappelle:

Even if you try and identity yourself as GoogleBot, you get nothing of the advantages unless connecting to dumb servers. Google bots -- as well as other crawler systems -- only have a well specified range of IP addresses from which they can operate an access remote content. No matter what kind of connection to the internet you have, chances are 100%, ie.:your provider won't let you, that your address won't match the specified range. Thus you'll get nothing, just be a stranger whom server admins look and say, "oh look at that n00b! He thinks he's google and can access out content!"

General ID spoofing (as, say, updating the current user agent string options) would be a good thing to have. But it can't help you masquerade as a crawler.

100% off topic !!!

This is not my aim but... hey ... Have I asked for comment on a possible aim? No! I think no!


See this option on firefox for a clearer view...

aid85: a feature without any use it simply bloat. that's why it is 100% _on_ topic to discuss if a feature would even provide any real usability improvement at all.

-1
UA spoofing is currently a necessary evil forced by poor quality web-design, but I look forward to the day when UA strings are used only to inform web sites of who has been visiting and spoofing is no longer of use to any one. In the mean time, I'd rather not proliferate the practice, especially if it's only to cheat web sites out of their subscription income.

opinions... no use for you may be usefull for others...

What do you want to do with UA spoofing is not of my interest...

In my wish list there is a UA spoofing like in firefox to do what everyones want to do... and nobody have to use it if there is it... but everybody could not use it if there isn't it...

Can you provide a practical use case where masquerading as Google Bot, from an options system integrated into the browser, that allows you to actually present a case if you talk about "opinions" and "things people do"?

"What do you want to do with IA spoofing" may be not my interest, but it is the developers interest as that is the material they need to take the decision whether to integrate the feature or not!

And for the record, the "UA spoofing like in Firefox" it has to be done with a plugin, whereas, in case you haven't searched the forums or the help site or any sort of Opera help, Opera already provides. It's just an issue that, last time I checked, the UA strings were kinda outdated...

Yeah sometimes the UA strings don't work because they're spoofing an old version of Firefox or IE

Originally posted by Astrophizz:

Yeah sometimes the UA strings don't work because they're spoofing an old version of Firefox or IE

The latest versions of Opera spoof IE 8 and Firefox 3.5

More in the past. Still, Firefox 3.5 is a bit old since 3.6 came out in January.

how to do it ?

opera:config#ISP|Id = googlebot/2.1
Need restart.

alleluja

some ontopic answer

(offtopitc: @RyanChappelle: can you provide me the aim to know an example of use that I or other can do with UA spoofing ? Water is water even if you want drink it warm it or drop it !)

I'll try opera:config#ISP|Id = googlebot/2.1

thanks all ontopic ;-)

The only use cases for Googlebot that comes to mind:

1.- testing pages with sitemap feature (sitemap.xml in root directory) or any other form of crawler-aware indexation system, to check that the crawler receives the correct/most updated version of the sitemap and can follow all links (it if couldn't, the site rank may drop due to 404'd links). It also helps to check that the text output that the crawler captures from the page is parseable/searchable, and thus usable in search results.

2.- news, journals and sale sites provide links with their products, articles or downloads free for Google to access and index so that they show in the search results, but when visiting with a normal browser, the content is privatized either behind a DRM or a registration/fee screen -- thus resulting in what amounts in law to fraudulent advertising (but why no one takes those people to court I don't know, since those institutions are defrauding both the audience and the search/ranking system). if the servers were dumb and didn't check for source IP address or crawling capabilities, simply masquerading as Googlebot would provide people with access to those privatized resources.

Note that the method above (customising the ISP value) does not allow to bypass any of those issues, as the crawler must be identified as Google by both the vendor and application strings.

does not allow to bypass any of those issues

Wrong. In reality, many sites only checks the existence of the string in the user-agent.

also:
ad 1) why should a normal browser not be able to access that xml file?
generally: if I check for web crawlers I'd also check the IP as identifying as a different browser is easy (would even be possible large-scale through a proxy that rewrites http requests), but spoofing the IP-Range is not.

↑It's not that a normal browser can't (because it can), but that masquerading as a search bot allows you to examine your is your server responding to the crawling, ie.: is an error returned for some reason, or something. Mostly for debugging purposes.

And I would really like to see some examples of servers with "private-advertised-as-public" content that only check for the UA id, just to have a fit of laughter when I point the finger to their sysadmins.

Test links.

If people are coming to this thread "needing" this functionality now - why not just use Proxomitron???

Originally posted by lucideer:

If people are coming to this thread "needing" this functionality now - why not just use Proxomitron???

Possibly the same reason some prefer to use quasi-builtin functionality for userjs management instead of a dll/exe.

Originally posted by fearphage:

Possibly the same reason some prefer to use quasi-builtin functionality for userjs management instead of a dll/exe.

? There isn't any built-in functionality though... is there?

Originally posted by lucideer:

Originally posted by fearphage:

Possibly the same reason some prefer to use quasi-builtin functionality for userjs management instead of a dll/exe.

? There isn't any built-in functionality though... is there?

I think that's his point.

Originally posted by lucideer:

Originally posted by fearphage:

Possibly the same reason some prefer to use quasi-builtin functionality for userjs management instead of a dll/exe.

? There isn't any built-in functionality though... is there?

I was speaking about the UserJS Manager Unite app actually. Some people prefer almost-native over external tools. I pointed out that there is no way to view the content of your stored databases and webstorage. The response was "use sqlite explorer". Again this should be native. I can view the content of cookies so why should databases be any different. Native is more convenient than 3rd party in many cases. I'll wait for the retort...

Originally posted by fearphage:

I'll wait for the retort

No retort. I at no point made the claim that using proxomitron or any external tool would be better than having this native. I was just supplying the OP and others with the information that external tools are available IF they really need this functionality in Opera now.

Btw - I couldn't possibly agree more with the thrust of your comment above - I almost always prefer native to 3rd-party, hence the graduated support of old "extensions" wish.

Originally posted by Lex1:

how to do it ?

opera:config#ISP|Id = googlebot/2.1
Need restart.

yes, but opera crash often with this option.

+1

Give us a sixth option to select a user generated UserAgent!

And please use different version number UA for masquerade; a different one from really existing Mozilla, FF or IE versions.

A former Opera version used identification as IE for default setting. I don't know how many opera users still use these old setting. Maybe several counters in the web interpret these browsers not as Opera. So maybe some ratings are not correct. Especially if 'Maskerade as ...' is used by default, I think no one will count this browser-request for Opera. Am I right?

I think editable user agents would be a good thing. Right now some sites don't work with Opera what it identifies as Opera. I do like being able to spoof as Firefox or IE8 but I think we should also be able to spoof as any browser we want such as Opera Mini, Google Chrome, Safari 5, etc...

Some websites (such as tax prep sites) run excellent in Opera but if your on linux you cannot use them. Some block anything but Windows (Or OSX) and being able to change your user agent while on such a site would be very beneficial.

+1

Please state a "white-hat" use-case for editable UA string

Originally posted by c69:

Please state a "white-hat" use-case for editable UA string

Testing for browser sniffing. In some cases browser sniffing even leads to a lack of access for untested browsers. It is patently ridiculous if a site supports Chrome but blocks Iron or supports Firefox but blocks Seamonkey. Sometimes when Opera is not tested by a site, lazy Web administrators will block Opera rather than allowing access. Barring JS differences or Opera bugs, the site most likely works, but the site does not bother in this case. Realistically the user agent should not matter, but that is not the case today unfortunately.

Turin: we already have the "Identify as/Mask as"-options for exactly that.

Originally posted by serious:

Turin: we already have the "Identify as/Mask as"-options for exactly that.

Which masks as an older version of other browsers and doesn't allow masking as Chrome or Safari.

It masks as IE 8 and Firefox 3.5.6 (you can check for yourself http://whatsmyuseragent.com/ ) which are pretty much the latest versions (okay, there is firefox 3.6, that's not what I'd call a version discrepancy that anybody should filter for in production)

Yeah I suppose so. Anyways it would be nice to have Chrome and Safari useragent strings.

Originally posted by Turin:

Originally posted by Testing for browser sniffing. In some cases browser sniffing even leads to a lack of access for untested browsers. It is patently ridiculous if a site supports Chrome but blocks Iron or supports Firefox but blocks Seamonkey. Sometimes when Opera is not tested by a site, lazy Web administrators will block Opera rather than allowing access. Barring JS differences or Opera bugs, the site most likely works, but the site does not bother in this case. Realistically the user agent should not matter, but that is not the case today unfortunately.[/quote:

Turin,
Well stated. I have known this for years, and been reporting it to Opera. The finally changed the UA string for Firefox, AFTER IT WAS ONLY YEARS Out-Of-Date!!

I vote for this, since Opera programmers refuse to check what is the current Firefox version and Gecko Engine when they are ready to release the next version -- whether Stable Release or Desktop Snapshot. Just a simple edit before compile would SAVE a lot of User Frustration. As Jimmy Buffett once said in song lyrics: "The answer was so simple, it just plumb avoided me".

Here's an example where the built-in UA strings fail: Facebook can use websockets for chat and I would like to use this in the latest Opera 10.70 build. However Facebook won't recognize the snapshot Opera as being able to use websocks, but neither Firefox 3.5 (nor Firefox 3.6) has websockets. Chrome has had websockets support since version 4 but the Opera UA strings do not include any version of Chrome.

Originally posted by Lex1:

how to do it ?

opera:config#ISP|Id = googlebot/2.1
Need restart.

That, as I have noticed, adds the ID to your user agent rather than change it. For example, I have Ubuntu 10.10 set as my ISP, and this is what my user agent reads:

Opera/9.80 (X11; Linux x86_64; U; Ubuntu 10.10; en) Presto/2.6.37 Version/10.70

As you can see, all what it did was add Ubuntu 10.10 to the user agent. Not the most effective method of changing your user agent, but if you're on Linux (like I am) you can identify your distribution in the user agent rather than let it be a generic distro. Having a completely configurable user agent would be nice in some situations (such as browser lockout through browser sniffing,) though.

This topic is duplication of earlier one. Please, if you want new feature - do not clone feature requests, but agree with older one. This is the only reason why Opera developers will satisfy any request here - because it is a huge heap of trash.

I have quite a few web sites that will not function when they see Opera. I can go to change the site individually but first I have to move away from the site clear the cache and cookies.

Thats quite a bit of work when this happens so often. There should be a global option. Maybe someone will create an extension to do this if the capability is there.

+1

hey this is one good feature

sorry to say that why i use Firefox 4

b'coze its very hard to download content of nokia ovi store from mobile
's costly internet

and any one can't download content of ovi store in pc
without changing useragent

but firefox makes it possible

so team opera add this feature

@rehanforall

no need user-agents for download from ovi store
just install this extensions Ovi-Store

@Crsmsg

hey dude ovi store is only example

can you tell me if want to download google search google maps and youtube apps from www.google.com

its imposible to download without changing useragent

and one thing more opera ovi store extension works website of store.ovi.com which is desktop site which load very slowly

but if you change useragent then you can browse store.ovi.mobi which is mobile site who loads faster then store.ovi.com desktop site

i think now you understand what we want

+1

+1

+1

Originally posted by Astrophizz:

Which masks as an older version of other browsers and doesn't allow masking as Chrome or Safari.

// There is no need for masking as Chrome or Safari as they use a similar user agent string to Mozilla Firefox. Very pointless to add them.

// Below is the user agent for Chrome, Safari, and Mozilla Firefox.

'Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US) AppleWebKit/534.13 (KHTML, like Gecko) Chrome/9.0.597.84 Safari/534.13'
'Mozilla/5.0 (Macintosh; U; Intel Mac OS X 10_6_6; fr-fr) AppleWebKit/533.19.4 (KHTML, like Gecko) Version/5.0.3 Safari/5'
'Mozilla/5.0 (Windows; U; Windows NT 6.1; de; rv:1.9.2.13) Gecko/20101203 Firefox/3.6.13'

Taken from http://whatsmyuseragent.com/

What I would like is ability to increase “privacy” of the user agent string, eg. set it so that instead of all the things it contains only “Opera/9.80”, or “lie” about version zeroing the last digit.

signed, if you take the end of the current UA string:

Presto/2.7.62 Version/11.01

imo it could easily be truncated to

Presto/2.7 Version/11

but imo that should be done by default, not by customizing the UA string. Though various other methods could be employed to eg. find the real microversion of the engine (feature and/or bug testing through javascript).

Aside: Actually I'm not sure why there is even both Engine and Browser version noted, wouldn't one be enough? I mean for any given browser version you know the inbuilt engine version as minor engine updates also result in a minor browser version update.

Originally posted by serious:

that should be done by default

Yes, I agree.

Originally posted by serious:

Though various other methods could be employed to eg. find the real microversion of the engine (feature and/or bug testing through javascript).

I'd guess that not to such granularity as with version number since two “minor” releases could have the very same features but one could have improved performance or what not.