#StandWithUkraine
Andrey “Rarst” Savchenko —  — Software, Web — browser, opera, server

Opera Unite – are browsers out of feature ideas?

opera_unite_icon With their bold claim of reinventing the web Opera had unveiled Unite service. I am amazed how some made absolutely accurate prediction while I was looking in opposite direction of cloud stuff.

By the way this marks browser innovation as dead. Let’s see where browsers go from here.

Presentation

Opera served their idea under slogan a Web server on the Web browser. Now how fast can you appreciate idea of running server as part of your browser? Took me some minutes.

For comparison Google presented their Wave project under slogan if email was invented today.

Feel the difference? Anyone would be interested in better email, but few are interested in running local server. And core question - does web server in browser make browser better?

Unite is actually standalone technology that (ab)uses browser as delivery platform.

Claim

Reinventing the web.

Running server locally is nothing new. Skim my blog only and you will find posts on how to run locally servers for HTTP, FTP, Direct Connect, UPnP and I can (and probably will) easily churn out few more on torrent and such.

Comparing to that Opera Unite is:

  • much easier to use;
  • much less powerful;
  • roughly equal in being toy for techies (at start?).

To change the web it lacks novelty and more importantly leverage.

Technology

Opera Unite is going to be embedded in future versions of Opera browser. It allows to serve content from your local computer to Internet and host Services (think rich web pages) with different functions.

It is mashup of several existing technologies:

Which means:

  • it is hardly powerful server, modern sites run via things like PHP and MySQL. No mention of language support beyond JavaScript or database of any kind in Unite (so far);
  • Opera widgets never got much love or adoption, at my work in office packed full of Opera users there is not a single widget in sight;
  • it is relatively open scheme but you still need proprietary account (shocking :) ) and locked in Opera online ecosystem.

opera_unite_interface

Usage

  • file-sharing – absolutely yes but on small scale. Demand for simple and efficient file sharing is huge. Whenever I put file in HFS or Dropbox and give direct link in IM people are often dead envious. Still there are heavy reasons for so many online services (RapidShare, etc) and specialized protocols (FTP, torrent) to exist and Unite can’t replace either.
  • collaboration – possibly. There are numerous approaches to this. Will be decided by API, sandboxed environment and if they will be enough for good and popular implementation.
  • social network – not happening. Many seem to perceive personal focus of Unite as basis for social-type platform. Imagine platform where you are neither visible or contactable when your browser is closed. There is no server component to Unite. It might integrate existing social networks but cannot become one.
  • contact point – possibly. Some instant messengers have live chat feature and I can see Unite acting as one. Being online is needed for real time conversation anyway and running it locally might be more flexible and customizable than third party services.
  • synchronization and cloud – Unite is anti-cloud. It is completely unfit for synchronization unless it will be extended to work with some kind of online storage.
  • something my brain is too small to imagine – possibly. Can you think of anything not mentioned?

Browser evolution

It is indicative how Unite steps sideways from browser functionality. By the way so does Firefox with numerous extensions that often have little to do with web. So does Chrome that puts performance with online apps before local features.

I believe that modern browser war is decided not by features but by marketing. There is no place left for browsers to evolve as browsers except in performance. And while performance is important frankly it became routine and expected.

So while Google took de-featuring approach with Chrome – Opera seems to look for an edge in building web services on local base, skipping need for own cloud and online services portfolio.

Competition

It is a given and beneficial that useful browser innovations are copied by competition. However Unite is browser feature like none before. Would other browsers be bold enough to pillage it?

  • direct implementation – even if Unite will be open enough to create compatible server code I believe no other browser will openly adopt specialized and branded protocols from competitor. Marketing hit will be too heavy.
  • idea implementation – easily possible but it is questionable who of developers will see same benefits and come to same conclusions as Opera Software.
  • in Internet Explorer – would be nothing short of awesome but sadly not happening. Microsoft has bit too much trouble even keeping simple browser bundled, not even mentioning constant struggle to patch holes in it.
  • in Firefox - attempt(s) will sure be made, either official or third party. Will it take off is another matter.
  • in Chrome – not happening. Google is pro-cloud, Unite is antagonist to that.
  • in Safari – who cares.

Overall

Opera Unite is boldest move in years for browser niche. It throws out of the window what browsers are supposed to do, as well as trend of cloud adoption. It is also completely risk free for developers and only takes little commitment of providing APIs and hosting for Unite Services, that will be mostly developed by willing third parties.

It is far from game changer but it is truly brilliant move that can still reap a lot. Move not by Opera browser but by Opera Software. As expected from I guess.

Link http://unite.opera.com/

Related Posts

13 Comments

  • Simakuutio #

    Just imagine this Unite feature in mobile devices... (yes, Opera have that technology already up & running, without Unite yet but I see it coming...). For example, Opera Mini + Unite... think what happens next? Or some other Mobile OS thing from Opera using Unite... Yes, this could change things a bit...

  • Rarst #

    @Simakuutio So what? Running server on mobile devices is nothing new (again), modern smartphones can do that. Of course mini might expand device base for this. Still, what would you personally serve from your phone? :) I can't think of anything at moment. Mobile OS from Opera is unrealistic. They don't have anything remotely close to it and no devices to use it on.

  • Geek Squeaks’ of the Week (#15) « What’s On My PC #

    [...] Opera Unite – are browsers out of feature ideas? [...]

  • Coreinsanity #

    Hey, its Core from ghacks ( Was the one to reply to you on the blog started there ). I agree with your comparison, really if you compare it to a current webserver (Which includes FTP/PHP/MySQL/Etc) it is relatively basic and a lot easier to use, sadly you can’t keep the power of something like a full on webserver and still have the easy to use dumb-it-down UI, or else this Unite might be interesting. Quite frankly, I don’t see this being used for much of anything aside from a real-time chat page (Ajax Chat type page), or a photo library like thing. You are too limited; if you want a blog you have no database on which to store your entries, you have no language on which to develop the backend code that would even process and store / get those entries, which means you also can’t use a file-based storage system. Personally, I don’t think this just makes social network stuff impossible; it makes it impossible to do much of anything aside from what I listed. Unless they expand on what they currently have, it won’t pick up big for anything aside from simple stuff. As you said, this won’t replace anything like FTP or Torrent, they are too good of systems. Torrent especially, as it less limits your speed based off of the upload speed of the server, but rather the upload speed of everyone sharing the file (For instance, EVE Online has their client downloadable via torrent, as do many Linux distributions, to provide an example for the people who think it’s just for illegal stuff; in fact World of Warcraft uses P2P application kind of like torrent for its patches and game download now, if I recall, and CCP with EVE is picking that up as well (though it’s optional on both)). As far as widgets go, I tried this browser a while ago, as I kept hearing about how good it was and about how it had widgets like Firefox, etc. Well, I tried it and found pretty much every widget they had to be useless to anything I do. In Firefox, I have all of about 9 Extensions installed; all of them EXTREMELY useful for what I do (Web Development), for instance, Firebug/Web Developer/MeasureIt/Some other stuff that adds to Firebug, as well as ad blocking stuff like AdBlock Plus/NoScript/Greasemonkey(Not really only for ads, but I use it as such to get more customized blocking on specific pages). The list of useful extensions for Firefox is massive, and I think they are that they are the key to it’s success. I will admit that this is an interesting move, and a bold one at that; but I also don’t see it being a game changer, or switching many people over to Opera. I simply wouldn’t give up the functionality and usefulness of Firefox for it. Now that I have gone and rambled off a book for a reply; I think I am going to go remake a better version of the real-time chat they have in PHP/Ajax/MySQL and all that good stuff. Should not take me but about a day. P.S. Nice blog site you got here :) Bookmarked it.

  • Rarst #

    @Coreinsanity Welcome and thanks for extensive comment. :) Yeah, most important question is how powerful Services could be coded for it. Unite will find some users in any case, that is guaranteed and part of foolproof plan worth admiration. But only scope and potential of Services will determine if it can get meaningfully large user base. Firefox clearly strives on extensions. Actually [too] many people using it start their story with "I need extension X". You was misinformed about Opera widgets. There are some useful ones but they are nowhere near Firefox extensions in either power or suggested usage. Power of Opera is very complete core package that offers full online experience (web/mail/rss/news/irc/torrent/whatever) out of the box. There are some great customization/personalization features as well but the way they implemented mostly caters to geeks.

  • Coreinsanity #

    @Rarst I was skeptical about how useful the widgets could possibly be when I heard about them; I figured since no one I knew used it (Aside from the one guy I talked to), it couldn’t have had that many really good widgets (Or they would have been using it). About the only thing I really like about Opera (Yes, bitter-sweat Opera comment incoming) is the UI. It is nice, simple which I like. Aside from that, Firefox is much, much better for me. For stuff like RSS, I would rather make my own PHP app to take care of that, and all that stuff. I don’t really use IRC so I wouldn’t want that built in (And if I did, I would honestly rather use mirc, then I would not have to have the browser open to talk on IRC). Also, on the note of browsers picking up the Unite type webserver thing; I think some will, most likely Firefox will try either with extensions or official release. Either way, I hope they add in a way to support PHP/MySQL/MSSQL/Etc if they can, as it would make it much more viable to be used for something. On the note of browsers adopting other features on other browsers: I actually think it’s quite funny, along with the coming of IE8, Microsoft actually implemented several useful tools; One of these tools was their “Web Developer” thing (I think it’s called). I looked at it, and I was actually quite surprised as to how much it looked like Firefox (Seriously, if you have not used both, I would recommend you do just to see how similar their UI and functionality is). I think it’s awesome IE has it, just kind of funny as to how much it looks like Firebug. Also, no problem for the extensive comment, I have a habit of writing long comments when I actually do reply to something. @Simakuutio Yea, I don’t really see anything I would host off of my phone, quite frankly. About the only thing I could think of would be if some one is traveling and feels the need to update the world on it via a mobile real-time picture gallery kind of thing, aside from that I can’t really see a point due to the fact that you can’t store a database (Which removes the possibility of any kind of mobile real-time blogging that they host), or anything like that. Really, they killed/are killing it by not having the added functionality of the current webservers; Even if the control over the webserver configuration there, stuff like PHP and MySQL are about as mandatory as it gets to running anything truly useful now, and honestly I don’t know how well a phone would even be able to run PHP or MySQL kind of blog (I have a simple phone, I have never been much into those expensive toys like the iphone, id rather spend my money on my computer).

  • Rarst #

    @Coreinsanity Actually Opera is also working on Firebug-like thingie. It's called Dragonfly (Tools > Advanced > Developer tools). Sadly it sucks. Maybe they will get there but I hadn't heard much about it since it was announced as alpha. I know that power users won't need or use most of internal stuff (I don't). But for regular users Opera is excellent choice. So is for portable usage. So is for keeping things synchronized and/or contained. It is just one of possible approaches and there are plenty people who like it. Same with Unite. It seems grossly underpowered but if I turn off geek part of brain for a second I see that many will love it for integration in Opera and simplicity of usage. It fits Opera. I guess it makes sense.

  • Coreinsanity #

    @Rarst Yea, it does fit. Don't get me wrong, Opera is not *bad*, I just would not pick it. I can see it having good uses for the average users that don't want to bother with installing extensions or any of that. Didn't know they are developing a Firebug-like utility, not surprised though honestly. But yea, for the average user it's great, but to me adding stuff that just simplifies something else just for the sake of adding simplicity is counter productive. If the simplicity makes it only useful to the average user, that's great. But unless they actually add alot more stuff to it, even the average users is going to be limited in what they can do. For instance, it would be great if they included PHP / MySQL support, and make it easy to install that support. Even more so would be if they added something like Fantastico (Or whatever its called) that comes in most cpanel hosting now. This would allow the actual backend developers like me, to do what I want with it with increased customization to the pages I make, and the average person to one-click install a blog that runs when their site is up, giving them the functionality of a blog but not needing the technical know-how of installing or making one. All they need to do is figure out how to configure the blog (Which is not hard to do, if they actually try). An added option to have Opera run the webservice in the background when the browser itself is not actually running would further reduce the limitations of it. They need to make it keep its simplicity, however increase its functionality - then it will have a much bigger impact on stuff, I think.

  • Rarst #

    @Coreinsanity Keep in mind that Unite is quite sandboxed for security reasons. It won't become anything close to regular web stack like that. I guess it would be wise to nap on Unite until Opera 10 release (at least). Personally I don't see myself developing any services for it (I am less than clueless in JS) so I can only expect to see what others will do with it. btw on running in background - Unite spiked feedback that says Tonido http://tonido.com/ is same thing minus Opera. :) Had bookmarked that one to play with.

  • Coreinsanity #

    @Rarst Exactly :) You can't have functionality without increased risk of less security, as there are more things to account for security wise. Though, if they could find a way to account for them and make it secure, it would be a fairly interesting service (One I still probably would not use, but it would become more widely adopted I think). See, I have the mindset where, if I can code it using PHP, or set it up on my computer, I will do it. I love coding random stuff in PHP (For instance, Ajax real-time chat system). Where a lot of other people would go off and use something like CodeIgniter, or Smarty, I have taken the steps and started working on a framework of my own for template loading and general functionality for me to build the rest of the website backend off of, so that we (Me and my Friend/Boss who I contract my work out to) can make websites that much quicker for the client, and the system is mine and I know how it works. I am good with Javascript (Not as good as I am with PHP, though), but I simply don't have the time to make my own Javascript framework for Ajax and all that - I use prototype (http://www.prototypejs.org/) I find it to be quite nice for what I need. It has some very good features if you ever look into Javascript but don't want to go through the trouble of making your own small framework to handle some stuff. But yea, as I said, Opera is not bad it's just doesn't quite fit what I need/want right now. Also, will have to look into that, it looks interesting. Thanks for the link.

  • Presentado Opera Unite - Foros Softonic #

    [...] [...]

  • Pallab #

    Wow! How the heck did that guy guess it correctly? We all knew if was something related to cloud computing but the idea of webserver in a browser didn't even remotely occur in my brain. Kudoes to Kas Thomas. And there are still plenty of features to implement. Opera needs still lacks stuff like form filler (not a password filler, but full fledged form filler like lastpass or roboform). It desperately needs to create an userjs manager as well as an userjs gallery.

  • Rarst #

    @Pallab Yeah, he made very accurate guess. Probably his train of thought coincided with that of Opera developers. As for forms I am used to filling those with Opera Wand and Notes. Not completely automated but works and is native. Never got to playing with JS stuff. It indeed could use better interface and some kind of catalogue.