Slashdot Mirror
Opera Unite is a Hail Mary
snydeq writes "Rather than view it as a game-changer, Fatal Exception's Neil McAllister sees Opera Unite as a Hail Mary bid for Opera to stay in the game. After all, in an era when even vending machines have Web servers on them, a Web server on the Web browser isn't really that groundbreaking. What Opera is attempting is to 'reintermediate' the Internet — 'directly linking people's personal computers together' by making them sign up for an account on Opera's servers and ensuring all of their exchanges pass through Opera's servers first. 'That's an effective way to get around technical difficulties like NAT firewalls, but more important, it makes Opera the intermediary in your social interactions — not Facebook, not MySpace, but Opera,' McAllister writes. In other words, Opera hopes to use social networking as a Trojan horse to put traditional apps back in charge."
Data Center Knowledge has a roundup that looks at some of the problems with this approach, including security issues related to running a server on a desktop app and bandwidth consumption. If your browser-hosted site gets busy, you think your ISP won't notice?
The summary makes it sound like Opera is making a last ditch effort to stay relevant, which is clearly not the case. Opera has always been in a dominant position in mobile browser marketshare.
Source
This will be a blast of a tool for web developers. Imagine developing your work anywhere on your laptop, regardless of availability of internet connection.
The main purpose to the servers that Unite can provide, is that they the most common type of computer connected to the Internet (one that does not have its own static IP, and cannot accept connections due to either a home router or a firewall) can act as servers. I've yet to find out much about the technical workings of Unite, but from what I can tell the main role Opera's servers perform is to allow the location of and the connection to a computer which would normally permit neither. Once a client has found and connected to a Unite user, does Opera still continue to act as an intermediary, in the same way a cloud service would?
If we can put a man on the moon, why can't we shoot people for Apollo-related non-sequiturs?
I released an open source web browser called SupraBrowser a while back. It has very similar characteristics to Opera Unite, in that it's designed to act as both a client and server at the same time (we called it a "servant") :).
This was more of a research project, as in fact, it was designed as a research and collaboration system for financial services companies and is currently used heavily by several very large financial services companies. It's almost like a combination of Google Wave and Opera Unite, in that it's based on a secure real-time messaging layer (xmpp/jabber wasn't stable or mature enough when we started....if we were doing it over today we might use jabber, but we also had the need for a lot of queuing and persistence that jabber wouldn't have provided), where all communication is completely encrypted using 3DES and a zero knowledge authentication. It supports email, mailing lists, group posting boards, link sharing, workflow, and a bunch of other really innovative features.
That said, I don't know how to manage an open source project and generate a community around our efforts other than posting to various blogs every once in a while when I see something related. Even still, its' frustrating because we actually went far down the road of trying to do kind of what Opera is doing, but without a middle man/trusted third party (hence the requirement for SRP Zero Knowledge auth). We want to build a personal cloud collaboration environment where data becomes user-centric and controlled, where other services federate from that single point of truth owned and controlled by the user.
Given that it's a research project, there are also some very innovative ideas, and I have yet to see anyone implement tagging better or provide a better way to manage personal information. I have over 25,000 bookmarks and files that are all full-text indexed (on Lucene), and tagged so that I can easily get back to stuff and correlate it within my existing cloud of data.
This I think is one of the real weak points of the open source model. If there is something very innovative, it generally requires sales and marketing to shove it down users' throats given their natural tendency to resist change. When the users are the developers are the users, the self selection process tends to restrict certain things. I can think of no other explanation for why releasing 4+ years of effort has been almost completely ignored. If someone can point out why the open source community has ignored SupraBrowser I would be all ears!
If anyone has any ideas or feedback, please reach out to me! suprasphere ____ @ ___ gmail.com
David
Yeah right. If Microsoft had done this first it would have been hailed as revolutionary.
My karma is not a Chameleon.
Opera has always tried to inovate, they have an excellent product, with lots of good features and I couldn't believe how fast the pages loaded in comparrison to firefox. Its the little things though that destroy its market share and preventing it from gaining more. In opera on my hardware middle click doesn't map to scrolling or even as any default middle button action. It never has in all the years it has been in development. I am always impressed then disappointed . I know if I wanted to I could probably find the problem, but how many of your non-tech friends would try to fix it - I know my DAD would just say " This doesn't work, where is my old browser?"
The author may be right in the sense that Opera is attempting to find a way to distinguish their product from the competition, but I think he's missing a few points.
There are many reasons why you might want to run a personal web server on your local machine. It can act as a proxy for example. Since it is fairly easy to program it with scripting languages it can do a lot of interesting things. Granted all that functionality could be built into the browser itself, but if you can tap into a lot of existing code and also create a more organized stack for this kind of thing it could be useful. You could do most of the things people use things like Greasemonkey for now, except probably better.
It could be highly useful for web app developers. With some specialized tools designed to help with things like AJAX debugging it could represent a significant draw. This is maybe not a huge market for the bigger browsers, but if Opera can get a bit of penetration into the dev tools market this way it could provide them with a new revenue stream.
It could be highly useful for collaborative web based applications which feature interactivity. For example it makes more sense to send a copy of every event the UI needs to process to a queue on the client side than to force repetitive performance-destroying polling across the net. Now the app need merely check a local queue using a local HTTP request, probably using AJAX. It could also be used to allow processing resources at the clients to be harnessed to do a lot of the work, possibly in parallel.
This is not a new concept, but nobody has really rolled out a useful version of it before. There are going to be issues like NAT firewalls etc, but there are various ways to approach solving them. Afterall, people play online games all the time that require them to open ports, etc for bi-directional communications. All this is doing is extending that capability to the web.
Personally I don't think it will catch on simply because Opera has too small a market share to make it worth people writing a lot of software that depends on it, but the concept itself is not bad. Perhaps Mozilla will experiment with this too, then it might go somewhere, finally.
"Malo periculosam, libertatem quam quietam servitutem." -- Jefferson
Never again will opera, football, and computers come together so succinctly.
I can't speak for everyone, but it's good enough for my family to share our photos for a few hours between other members of the family. No more need to setup a permanent account on Flickr, just point the folder where your photos are located and it works (there are still a few bugs to iron out but it's less than beta). It's easy for my relatives, we'll keep on using it as long as it's available, and I hope more "widgets" will be written to do more stuff with it.
ensuring all of their exchanges pass through Opera's servers first. 'That's an effective way to get around technical difficulties like NAT firewalls,
Well, ever since broadband came in people could run home servers if they want - OS X comes with a built in web server and the world hasn't ended. Lots of NAS boxes today include click-and-drool webservers and you can get dynamic DNS if you don't have a fixed addresss.
However - I've got a better idea: why don't they just store the stuff the users want to share on the central servers? I mean, hard disc space is about fsck all per megabyte these days, the servers can run 24/7, have a super-fast connection to teh interweb (not an ADSL line with lousy upload speed) and have the latest security patches applied daily by dusky, nubile virgins (well, 1 out of 3 ain't bad). Even if the server does get hacked then it doesn't affect the end user. Much better than leaving your PC on all the time, or having someone suddenly trying to download a video when you're in the middle of a networked deathmatch...
Then there would be loads of material on the servers, so people would actually want to visit them. Hey, they could even attach comments and stuff to people's photos, videos, news articles and things to say whether they liked them.
You could call it MyCRT, FlipR, ArseBook or ColonPling or something...
Should I patent this, perhaps?
In a survey of 100 programmers, 111111 thought that duck-typing was a good idea.
Yep, I'll bet that a profitable browser company that continues to expand and make more money year after year is definitely in trouble. People have been predicting the death of Opera Software for over a decade now, and yet they're still making huge waves in the internet market. When will people start to realise that you don't need to be the most popular product on the market to be successful? The browser market is absolutely huge, remember. Even 1% of the entire market is millions of satisfied users. The only thing that matters is that open standards are implemented, then we all win.
Think American football. A desperate throw to try to make a touchdown from an area of the field where you should really be focused on gaining a first down.
For examples, see any football movie involving a slow-motion throw in the last second of the game from too far away that the main character catches against all odds to win the game for Sunnyville High (or whatever) with cheesy music playing in the background.
Is this Slashdot? Is this the Slashdot that use to have every user complaining about having their files and information being held in "The Cloud"? The place where people use to complain about the security and privacy of their pictures if they were on Flickr,their Email and their Docs if they were all located on Googles servers??
What happen to all the 'get of my lawn' types that said, "I'll never put my information in the 'Cloud'. They can take my physical hard drives when they pry them from my cold dead fingers!!"
Now you are provided with (one of many) alternatives to have your files on YOUR computer AND the advantages of them being in the cloud (like you can access your files no matter what location your at and be able to share files with other people)
This seems like a case of you can't please any of the people any of the time kinda thing.
Back in 2000 Netscape did a despo gamble like this and its implementation of some java classes was bad. It allowed websites to create classes derived from the server side of the browser and access all the info in the hard disk.
Google for Netscape and Brown Orifice for more details.
http://www.securityfocus.com/news/70
These were Java bugs from 2000, not something Netscape intentionally allowed. A desperate gamble, WTF?
Such a security hole is waiting to happen. It is really a dumb idea from Apple. One of the biggest plus point of MacOS is that, it is safe and it does not have vulnerabilities. To put that reputation at risk by allowing the browser to dish out data to the outside world is really really a dumb idea.
Yes, there are security features. Yes there are things the user must enable for it to work. Despite all this, having server code loaded up in the memory of a browser is stupid.
From Apple? Who is Apple? Opera? Are you lost? It was Apple's idea? WTF?
Have /. mods gone completely fucking insane?
I feel that most people here is Slashdot didn't get Opera Unite:
address most issues people discusss here and elsewhere.
I have to say, the only thing less likely to be understood on /. than a religion metaphor is a sports metaphor.
Give me Classic Slashdot or give me death!
The fact they can provide basic DNS services free, subsidized by the few who use their relatively low priced premium services, shows even $19/yr is much higher than it needs to be. DNS probably costs the registrars less than $5 for most home user's requirements.