Submitted by “The Mugshot team has been off learning what we can about all kinds of not-traditionally-open-source technologies and audiences, but recently we’ve started thinking about how to roll the knowledge back into the Linux desktop. Here’s an update (all heavily biased by my personal point of view).”
People who avoid using online applications for whatever reason
I have nothing against online applications per se. But I _am_ reluctant to trust “free” services with a lot of my personal information.
So I hope they integrate p2p and other more decentralized services into their vision and not just all those shiny AJAX sites. Besides centralized web services feel a little anti free software. I like services I can study and modify to my needs.
My problem with online services is that the majority of them don’t let you download back your own content. For example, if Blogsome.com dies, I will lose my blog and all the content that’s there. That’s how I lost my pictures in the previous service I had signed up. They went belly up, and they didn’t even offer us to download our content before closing down.
”
I have nothing against online applications per se. But I _am_ reluctant to trust “free” services with a lot of my personal information. ”
Well you are not providing much personal information. You don’t need to give your passwords or anything to Mugshot. You can just give the username/identity name and it aggregates the _already_ available information more readily. That’s all.
“Besides centralized web services feel a little anti free software. I like services I can study and modify to my needs.”
Mugshot is completely Free and open source software. That client as well as the web services.
http://developer.mugshot.org/wiki/
“I have nothing against online applications per se. But I _am_ reluctant to trust “free” services with a lot of my personal information.”
But…
“So I hope they integrate p2p and other more decentralized services into their vision”
…you’d be happy to share your personal information with complete strangers?
“””
…you’d be happy to share your personal information with complete strangers?
“””
I would trust a P2P based service with, say, my love letters (not that I ever send or receive any), as long as the protocol supported gpg.
They could exist in a virtual sort of way as encrypted fragments. Something like my love life itself, I suppose.
Edited 2007-04-04 05:05
I meant P2P in the sense that if I run my own SMTP-server that is P2P-mailing, but if I use GMail then its centralized.
In the P2P case I’m in control of what information I give to the world. I can encrypt communications and storage, rout through TOR or what ever I feel like.
In the GMail case Google keeps track of as much information they can about me, who I know, who I mail, what I mail about and so forth.
All it takes is a strange SPAM or some not so clever guestbook entry one subpoena from FBI and I might find myself going to Guantanamo…
Edited 2007-04-04 15:59 UTC
You guys tried to build a Networked Object Model Environment once before, and that didn’t exactly turn out as planned.
Remember when you ripped out all the Network Object Model part of it because it was poorly designed, and too hard to implement effectively?
Whats changed?
”
Remember when you ripped out all the Network Object Model part of it because it was poorly designed, and too hard to implement effectively?
Whats changed?”
Well, nothing. KDE ripped out artsd because the sound deamon didn’t have any maintainer despite showing great potential during the early days. Linux VM was completely rewritten for Linux 2.6 among other things. Windows has so many revisions of COM like technology that everybody has lost count today.
So, no, nothing has really changed. Ximian thought Bonobo was good. Apparently it turned out to be too complex. DBus from Red Hat replaced some of the functionality and Bonobo is deprecated being ripped out. Happens all the time.
Also there is http://www.j5live.com/?p=348
Ximian thought Bonobo was good. Apparently it turned out to be too complex. DBus from Red Hat
I have no idea why you felt the need to insert Ximian and Red Hat into the picture, and as if only the two of them were involved in ORBit/Bonobo/DBus developments.
What DBus mostly replaces is ORBit, which was developed by engineers from Red Hat as well as a lot of people not affiliated with any company (though several of them later came to work on a new company called Helix Code).
As for Bonobo, it was also started even before Helix Code. The initial entries on the ChangeLog for version 0.11 (the oldest available on gnome.org) are from Feb/1999, and Helix Code was founded only in October of that year.
D-Bus is a technology that “fixes” a core piece of the GNOME platform. The mistakes it learns from were made by Red Hat, Ximian and hundreds of other independent developers that contributed to GNOME.
> The mistakes it learns from were made by Red Hat,
> Ximian and hundreds of other independent developers
> that contributed to GNOME.
not to mention the successes of other projects that tried the approach d-bus takes (e.g. dcop).
this is one of the strengths of the open source model: we can experiment and see which actually works.
Appears like Redhat is trying to join the dark side, by trying to steal your personal data now. Shame on you Havoc for supporting this. I considered you to be a sane person.
taschenorakel, please get a clue. Havoc simply tries to “get there” first. Others are trying to do the same thing too (e.g. Google). So you can’t say that Red Hat or Havoc support something evil because then the whole technology sector is evil. And if you have a problem with the whole tech, I suggest you go live in a cave.
Where Red Hat can make a difference is that they won’t lock your data away, but they will be an “open data” company. And that would be a good thing because it would offer more flexibility to the user of Google apps, or Live.com or Flickr.
The time of desktop environments has passed. Users don’t care whether it is KDE or Gnome. They don’t care which toolkit is used. they don’t want Nat Friedman to pollute their desktop with corporate choices.
the real enemy are these guys, who – for ideological resons – inhibit interoperability on Linux desktops. We really don’t care what toolkit is used to render the user interface. We just want it to play well with each other. We want to use Abiword with KDE and K3B on Gnome.
Plattform ideologists and lack of desktop standardisation are the only remaining obstacles.
It’s good to see Havoc Pennington thinking out loud again. Would this be a direct result of RedHats ambition to return to the desktop?
Somehow I get this deja-vu feeling – haven’t we had this before somewhere .
It reminds me of thin clients – although he explicitly says thin client are a very different audience .
Im not sure the web2 is yet as big it sounds in this blog .
Trusting external sources for potentially all tasks on the desktop – hmmm – :
x their servers can die
x the company can collapse
x they can change some liscence
x they can give away data
x they can change their mind e.g restricting access to your own data (imagine DRM on your own private data)
x things can just suddenly get lost (Gmail)
x it all requires net bandwidth
x more & more in the online world – there is a real one
IMO this online desktop will be lightweight desktop tasks (like google spreadsheet) & stuff which is online social stuff .
All these web2 services are free – right ?
If everybody uses them – they cant be forever – they could make additional features for a fee/subscription available .
Some data cant be trusted to just sit on someone else’s server with millions of others – well I guess mentioned in the blog – business desktops not included .
Some tasks just eat up way too much rescources to “run” online – photo & video editing .
Although leasing RAM & processing power of powerful systems available online would be interesting in such cases .
Kind of reminds me of supercomputers & exactly such allocation for projects .
I guess if it all had/has security & reliability in mind – then I could have more online presence via my desktop .
For online applications with the ever increasing processing power & RAM increase out there sexy flash based apps & general web programming brought to the desktop capabilities could be interesting .
Reminds me of widgets .. well i meant of course gdesklets .
“time of desktop environments has passed”
Well in the end the idea is to be productive – wherever the application executable is .
This GNOME-OD seems rather perfect for these little mobile computers .
People would use them more as a data flow controllers with a speech & video abilities – than to use its own processing power .
Music – phone – video (video on demand ? wasn’t that something already years ago of some kind ?)(video streamed to TV or the magic wallpaper that is a display as well 8) – word processing – general sharing stuff could all be done via this online way .
I remain skeptical 8)
Just IMO
Hi Guys
I don’t normally get involved in these sorts of discussion but simply had to make you guys aware that a lot of your theories are already blown or affirmed (depending on your standpoint) by desktopondemand.com
They’ve recently launched the kind of online desktop you guys are discussing and have for the most part managed to deliver just whats needed for now.
I’ve certainly been impressed by what they have managed to achieve (and I have used my fair share of thin client/networking computing environments).
Might be worth you guys checking them out as some of your viewpoints might change