“Xcerion is a Swedish Internet startup whose founders include ex-Microsoft employees Lou Perazzoli and John Connors. The company will make headlines later this year when they officially unveil what they call an “Internet OS” dubbed XIOS that runs in a web browser. We took an early look at the XIOS concept and had the chance to talk about the project with the company’s CEO, Daniel Arthursson.”
Dreaming in the ‘Cloud’ with the XIOS Web OS
About The Author
Eugenia Loli
Ex-programmer, ex-editor in chief at OSNews.com, now a visual artist/filmmaker.
Follow me on Twitter @EugeniaLoli
It’s a bit too early start developing a pure web based OS. While it’s an interesting idea, at the moment we simply don’t have sufficient standardization of various web standards. The standards that are being developed tend to be marginal and used only by in special cases or by small group of people.
IMHO, a web based OS would make much more sense when various microformats become standardized and we have common APIs for things like photo sharing, social networks and other common web applications.
While backend protocols (like POP3) could be employed, this isn’t good enough. POP3 is good for certain things but it’s quite limited. For instance, it does not include a spec for labels. How I wish there was an email client that could auto sync with gmail while understanding labels. For exmaple, if I label a certain email through the client, the next time I log onto gmail, the same mail would be labeled.
Currently there seems to be no API/protocol to achieve this. And this is just gmail, what about all the web applications out there? I am sure people who use a lot of social networking websites could give even more examples where interoperability and sync functions still need a lot of work.
We really do need a new generation of open/feature rich protocols.
Microformats are not the solution (to anything pretty much). If this ‘cloud os’ is based on true XHTML, then the current Microformats will be obsolete.
As far as I can tell from the very limited info available here, then cloudos will be based around a group of specific, integrated web applications. If this is true, then standards don’t matter (i.e. whatever formats are chosen will be automatically standard across the cloud os applications).
you could claim that it is too early to develop Office applications because there is no proper standardization of formats across office suites.
finally, POP3 isn’t really very limited. the whole idea of MIME headers is that they can be used for any key: value pairs of data required. There is no agreed spec for tagging MIME messages with labels, but there is nothing to stop a mail client using a
Label: xxxxxx
header to transmit data using POP/SMTP. So GMail might not understand the header, but that’s what the google mail API is for (there IS an API to achieve this). I’m sure that it would be trivial to write a thunderbird extension to synchronize offline mail labels with gmail labels.
In a way I agree with you that specs/protocols are a bit messy at the moment, but waiting for them to resolve themselves is not a productive solution. The protocols will be resolved by usage, not inaction.
[/rambling argument]
IMHO, a web based OS would make much more sense when various microformats become standardized and we have common APIs for things like photo sharing, social networks and other common web applications.
You mean, a web based OS would make sense if you owned Google.
Edited 2007-04-10 00:28
I think that the crux of the problem is that the people with enough money/manpower to accomplish that are also the people who are currently more interested in building their own little software fiefdoms. So rather than open standards/protocols being improved, we get “vertically-integrated” stuff like the Outlook + Exchange platform.
I’m a sceptic, so what I see here is a load of vapour-ware. All the articles (stemming from one press-release I guess) use phrases like ‘it plans to deliver’ and ‘will launch’ and I see no screenshots or even formal website.
I hope that something good comes of this, but until I see something solid, I won’t be holding my breath. ‘Specially as it seems to be based on XML
.
Edit: So it’s late here in the UK, I just found the website link. Still no solid info though.
Edited 2007-04-10 00:19
But this is nothing more than a web app not a web”OS” as presented. To run within a web browser, you need an OS underlying it with networking stack, etc. A true WebOS would be the total package… system of API’S, drivers, net stack, apps, etc. Looks to me like an impossible dream. If presented like it is, Why would I want to DL, much less purchase it?! I already pay the the MS tax, OSS Tax, etc. It’s impractical IMHO.
OSS “Tax”? You mean subsidy right?
Edited: Spelling.
Edited 2007-04-10 00:55
I think you might be able to call it a webOS if a full TCP networking stack were included in the BIOS, but then it wouldn’t be quite as “ubiquitous.”
Fair enough.
“Microformats are not the solution (to anything pretty much). If this ‘cloud os’ is based on true XHTML, then the current Microformats will be obsolete.
As far as I can tell from the very limited info available here, then cloudos will be based around a group of specific, integrated web applications. If this is true, then standards don’t matter (i.e. whatever formats are chosen will be automatically standard across the cloud os applications).”
However, I fail to see what’s the point of developing a new web interaction system from the ground up. Any web OS that wants to succeed will have to take in account current web applications and the standards they use. No one will use this web OS if it can’t work with current web applications, to do this we need detailed standards.
BTW, would you by any chance know of anything about how XHTML would make microformats obsolete? Maybe I misunderstood what microformats are, what I am looking for is the ability to say add an event (from any website) to any calender (that support the right standard) with one click of a button.
“finally, POP3 isn’t really very limited. the whole idea of MIME headers is that they can be used for any key: value pairs of data required. There is no agreed spec for tagging MIME messages with labels, but there is nothing to stop a mail client using a
Label: xxxxxx
header to transmit data using POP/SMTP. So GMail might not understand the header, but that’s what the google mail API is for (there IS an API to achieve this). I’m sure that it would be trivial to write a thunderbird extension to synchronize offline mail labels with gmail labels. ”
Hmmm, but there is no way for POP3 to sync the client with the web application, is there? I guess you could implement my “labeled syncing” by using IMAP + MIME labellings.
EDIT: sorry I was replying to stestagg’s post, but for some reason, clicking on the reply button didn’t work. This is annoying.
Edited 2007-04-10 00:56
Because XML namespaces (which can be utilised by XHTML) are a much more natural and arguably more correct way of expressing the information that currently gets shoehorned into HTML elements by way of Microformats at present. Despite the claims that microformats are ‘semantic HTML/XHTML’, they’re not nearly as semantically-rich as using a purpose-built XML-based format, which is what XHTML is designed to allow. In other words, microformats, although mentioning XHTML, aren’t actually built for it any more than many web browsers understand XHTML as ‘HTML expressed slightly differently’.
Isn’t this a little bit like Rebol? Even though Rebol isn’t run from XML but it’s own byte code similar to Java…
http://www.rebol.com/
How does this make any sense?
An OS is needed to run the web browser that in turn runs this web OS (actually, online operating environment).
How is this better than using VNC or w/e to run your desktops remotely?
1. Apps run locally and can run disconnected.
2. Data can be stored locally and online.
3. I assume the apps will just use port 80. VNC, etc require ports that are often not open on very secure firewalls.
Calling “OS” to a set of “applets” running on top of a web application seems to be very arrogant and ambitious!
An OS handles a lot of underlying things like thread scheduling, memory management, process isolation and communication, virtual memory, device management, etc. All those features are the most important part of an OS. The UI and all the user applications are possible because the OS creates an environment for that.
So, web operating ENVIRONMENT should be a more proper name.
Its pointless.
You boot an os, just to get a web browser, to connect to an online operating system.. so you can.. ? check your webmail..?
Why not just get a cheap flash drive and run portable apps (www.portableapps.com) that way no matter what comp you happen to be at, your apps work.
The whole idea of an online os just seems wierd..online web based email sure..
*cough* useless *cough*
I’m sorry, but this isn’t an OS. You could call it a framework, a desktop environment etc. but it’s not an OS. It has no connection to the hardware.
If you want a thin client, then use a thin client. This whole idea of making everything run remotely but require a fairly powerful local computer to run the display is crazy.
I can’t wait for people to get past this stupid web application fad and on to real remote computing.
Come on this is osnews, if it’s not an OS then why is the article up anyway? Personally i would not trust a WebOS for gaming since flashgames arnt that great and can be played on any PC with a decent webbrowser
i am gona take a different approach to this then other people here and start by saying i am excited abotu the project. In my work i travel to lots of different places, use lots of different systems, and dont always have one i like with all my stuff. i used to carry around bootable OS’s frmo flash drives but thats more work then i wanted to put in. it would be ncie to just go to any web ready pc with an os installed and pull up a common “desktop.” for those who asy it is no an operating system. isnt an operating system at its base level just a system that operates
.
I’m also excited. I use many PC’s and many OS’s. In reality, most of my computer use boils down to Firefox, Word Processing/Text Editing [Not the same thing] and code development [specialized text processing, may require some gui building].
If the “Web OS” can do these things well, I’m done.
Having said that, I think a more appropriate term would be Web Platform. That’s not as sexy as Web OS, but probably more accurate. Using this term, we are now comparing Ajax, Java, .Net, etc.
Nevertheless, while the technology may not be new, the implementation can still be revolutionary. By including Online/Offline usage AND storage, combined with optimized performance and polish, something seemingly new could be delivered to users.
The question is, isn’t this what Google is already doing?
[Edit: spelling, reduced size of quote]
Edited 2007-04-10 15:41 UTC