Weekly Edition
Return to the Press page
| |||||||||||||
KDE 4 promises radical changes to the free desktop (NewsForge)
NewsForge looks
ahead to KDE 4. "Its developers see KDE 4 as a chance to
experiment and introduce new concepts and applications that do more than
build on the strength of KDE's existing architecture. Just as KDE 3 brought
major transformations in that architecture, developers are looking to KDE 4
to transform the desktop experience and enable a surge in third-party
application development. With a KDE 4 release not likely to happen for at
least another year, the developers have plenty of time to
experiment."
(Log in to post comments)
KDE 4 promises radical changes to the free desktop (NewsForge) Posted Sep 27, 2005 0:54 UTC (Tue) by cantsin (guest, #4420) [Link] As an anti-desktop user, I best like the idea of the software layer that turns plain Qt programs into KDE programs. Given the stated intention that future KDE software should be programmed against the Qt API only, not the KDE API, it would also mean that future KDE software may be run in plain QT mode without the bulky KDE middleware.
does the desktop matter anymore? Posted Sep 27, 2005 4:32 UTC (Tue) by b7j0c (subscriber, #27559) [Link] my desktop hasn't changed much over the years - a virtual-window wm, a browser, a mail client, and a terminal. how much easier will kde4 make it for me to launch my browser? how will emacs suddenly evolve in the new desktop? seems more and more, functionality is going straight through the browser stack (http->x/html->css/js/dom->user) and bypassing the desktop app layer entirely. i see new desktops such as vista and osx etc touting operations such as photo management etc...hello? don't people use tools like flickr for this? are there people who really only store their photos on their local box?
does the desktop matter anymore? Posted Sep 27, 2005 4:53 UTC (Tue) by yokem_55 (subscriber, #10498) [Link] For people with literally thousands of photos, and with sufficientlylimited upstream bandwidth, the answer is, online is for sharing photo's with family/friends, but the main repository is on the pc. Webpage based services can do a lot, but can really only go so far....
does the desktop matter anymore? Posted Sep 27, 2005 6:01 UTC (Tue) by rqosa (subscriber, #24136) [Link] KDE isn't just a "desktop", it's a set of APIs, e.g. KParts, KIO, DCOP, etc. IMO, a better name for it would be "K Development Environment".
Flikr has some fairly strict size limits Posted Sep 27, 2005 7:46 UTC (Tue) by leonbrooks (guest, #1494) [Link] They won't take my 2000x1600 images, let alone my sister-in-law's 3504x2336 images from her Canon 20D. If I don't mind being unable to catgorise them myself, I can always PD a photo and send it to BurningWell.
does the desktop matter anymore? Posted Sep 27, 2005 9:09 UTC (Tue) by eru (subscriber, #2753) [Link] Indeed. I find that at work my KDE desktop is entirely covered with editors, shell windows, browsers and the like. The only features that see regular use are launching applications and flipping workspaces to allow launching even more applications, and to switch between them. So a much simpler system would be quite sufficient. For me, the only real advantage KDE has over fvwm and other classic WM:s is ease of configuration and adding desktop icons, buttons and other such widgets for launching apps...
does the desktop matter anymore? Posted Sep 29, 2005 8:56 UTC (Thu) by kdart (guest, #486) [Link] Exactly. When I assessed what applications I actually use, I find they all use the GTK toolkit, if that. I changed to a more light-weight setup and it works great for me:
XFCE - (menus)
All major desktop functions are covered here. I did try KDE in the past. The APIs look cool, but the actual apps are still lacking. The Kmail program twice trashed my home directory via an IMAP server!
does the desktop matter anymore? Posted Sep 27, 2005 9:27 UTC (Tue) by oseemann (subscriber, #6687) [Link] are there people who really only store their photos on their local box?
why would anyone want to entrust all their private memories to a foreign company? and isn't it tedious to manage thousands of pictures through a remote web interface? do they even accept my 20gb without me paying any money? isn't it much more convenient to have a smart photo management system on your desktop? without all that network latency and bandwidth restrictions? without storage restrictions and easy removable media support as well as backup&restore functionality? flickr&co are good for presenting selected images to a wider audience, but not for primary picture storage.
does the desktop matter anymore? Posted Sep 27, 2005 12:10 UTC (Tue) by pjs (guest, #10927) [Link] For most home PCs, a website maintained by a company is more reliable long-term storage than their local hard drive. Yes, it sounds crazy. In fact, about as crazy as never making backups. But that is the condition of most home computing, that backups are never, ever made. Not weekly, not monthly, not anually, not ever.
Virtually all home computing is done on Microsoft Windows. The most common "application" is entertainment. Often that's playing games, sometimes playing music or movies, other times it's trying nifty new software that amounts the resource-hogging eye candy. New applications (if you can go so far as to call them applications) are installed on a whim with little regard for long term system stability. Even without a buildup of malware, the system becomes less and less responsive as more and more, often poorly written eye-candy software hooks itself into various OS functions. Sometimes, a hardware failure occurs, though I've seen cases where home users mistake software problems for hardware failure (in which case they often believe recovering and saving their data is impossible). Human frustration also plays a significant factor.
Eventually, either a computer savvy friend does a clean install, or more often, a new computer is purchased to replace the old one which no longer functions well. How does typical home computer data get migrated to the new system? Most often, it simply doesn't.
By the time frustration with the computer's poor or flakey performance builds to the point of buying a replacement system, such a feeling of disgust and resignation builds that no attempt is made to save data.
Most home users, and in fact most business computer users don't have much awareness of where in the directory heirarchy they are storing their files. In recent times the situation has improved somewhat, with photo program defaulting to save photos in a "My Photos" directory, music applications defaulting to save files in "My Music", and so on. But if the files are located somewhere else, usually because some software had a default working directory configured elsewhere, only savvy users will have some idea where their data is actually stored. Many "user friendly" programs are designed for home users who can't understand the filesystem heirarchy and attempt to shield users from such difficult concepts, resulting in no opportunity to learn where the files are really stored in the ordinary course of using the program.
Even when they do know where their files are located, the task of transfering their data to a new PC, or backing up their data before a reinstall and restoring it afterward is quite daunting. Perhaps some home users can successfully network two computers and copy files (in the case where a complete 2nd system including a second monitor is purchased). But the reality is that burning files to cdrom is about the only managable way (for users savvy enough to even know where their files are) to transfer them between PCs. Of course, on top of knowing where files are located, copying to cdroms requires a PC still functioning well enough to actually burn a cd, and patience and thoughness to actually manages the process of dividing up many files between mulitple discs, patiently burning each one, and double checking to make sure everything important has been saved properly. It's an error prone process to manually back up to small size removable media, when any attempt is made at all.
For people who don't invest the time and money required for regular backups, who aren't savvy to the actual location and format of their data files, whose primary PC application is entertainment, who value new eye-candy over long-term system stability, and who may not always excersize caution and dilligence against infection of malware, long-term storage of data on external website makes a lot of sense.
does the desktop matter anymore? Posted Sep 27, 2005 12:27 UTC (Tue) by mrshiny (subscriber, #4266) [Link] Your view of computer users is pretty harsh, but, even if all you say is true, having worked at a computer store in the past I can say that most people who care about data on their old computer find a way to get the data off (assuming the hard disk is ok). If they can't get it off they call their local whizz-kid. If he can't get the data they send the data to a repair shop. My store had a repair shop where we often transferred a user's data to a new computer. In fact this was sometimes negotiated right into the sale of the computer. Many people don't know how their hard disk works but they can usually tell where there photos are stored, if they have taken any number of photos worth transferring to a new computer.
does the desktop matter anymore? Posted Sep 28, 2005 7:15 UTC (Wed) by Wol (guest, #4433) [Link] The trouble with asking the "local whizz-kid" is even that screws up often enough, thanks to Windows :-(
I backed up my brother-in-law's pc before wiping his hard disk. At least, I thought I had. It backed up everything fine without errors, except (1) it ignored hidden folders, and (2) because I backed it up from the top of C:, it ignored "Documents and Settings", because it was a hidden folder!
I didn't find that out until too late :-(
Cheers,
does the desktop matter anymore? Posted Sep 28, 2005 11:45 UTC (Wed) by carcassonne (guest, #31569) [Link] "except (1) it ignored hidden folders, and (2) because I backed it up from the top of C:, it ignored "Documents and Settings", because it was a hidden folder!"
Did you do a complete system restore from the backup, including all system files needed to run the OS ? Why no only re-install the OS for its distribution media and then only restore the user data ?
C: has to be the stupidest invention in computer history. No way to simply backup all of a user's data files in one sweep.
On Linux, the user will not save his/her data in /usr/sbin/. It'll be in his/her home directory and that's it. Want to save some system settigns, then backup /etc and that's it. Re-install system from distribution media and re-install the user's data files and possibly tweak some config files, but that's could really be not necessary.
I'd like to see Microsoft confined in 10 years from now publishing exclusively OSes and softwares for game boxes while the home PC would be the serious matter. People would have to power up their game boxes to run MS Office, which is not that bad after all, metaphorically speaking ;-)
does the desktop matter anymore? Posted Sep 27, 2005 15:07 UTC (Tue) by mmarq (guest, #2332) [Link] " But the reality is that burning files to cdrom is about the only managable way (for users savvy enough to even know where their files are) to transfer them between PCs. "
The only ?... savvy?... you mean here careless, which everybody is, even top scientists, or savvy is the not so clueless & dumb common joe/jane ?
All in all, it seams by your exposition that there is a dire need for Big, Fat, Bloated management features on all those desktops for helping users not be so careless and carry all those needed management tasks automatically and transparently(with voice Output/input also,perhaps) out of the hands of the lazy users.
Neither windows or KDE or Gnome is even close to that yet!
does the desktop matter anymore? Posted Sep 27, 2005 16:13 UTC (Tue) by b7j0c (subscriber, #27559) [Link] >> why would anyone want to entrust all their private memories to a foreign company?
seen many comments like this, of course no matter what point is made in a discussion, someone will rebut, but i am amazed that people are trying to pass off as insightful what in essence is a statement of fully trusting their own hard-drive to never fail as their one source of data. and now of course there will be a dozen comments about elaborate backup schemes like cds and dvds, but lets be honest, you don't do that either.
does the desktop matter anymore? Posted Sep 28, 2005 7:07 UTC (Wed) by niner (subscriber, #26151) [Link] > but i am amazed that people are trying to pass off as insightful what in essence is a statement of fully trusting their own hard-drive to never fail as their one source of data
Frankly I'd rather lose certain pictures of my girlfriend to to a faulty hard drive than giving them to _any_one else and I'm sure she feels all the same.
Regardless, I have my important data on the desktop PC and my notebook, so one crashed harddrive wouldn't be that much a problem.
does the desktop matter anymore? Posted Oct 4, 2005 10:01 UTC (Tue) by ekj (guest, #1524) [Link] cds and dvds are indeed elaborate. Especially for those of us with large data-collections. Every time I press the shutter-release on my camera I generate 3-10MB of data (depending on if shoot raw or jpeg) adds up quickly.Making atleast minimal backups are however *not* elaborate. Unless you think that clicking a *single* mouseclick once a week or whenever you feel like it is a lot of work. It doesn't take more than that with a separate harddisk and rsync (or any of the other programs for such stuff). Yes it costs like a hundred or even 150$ extra when you buy the machine to hqave two separate disks in there. If your data ain't worth that much to you, don't bother. I'm fully aware that two disks can die at once, or that the house can burn and so on. It's not fault-prrof, but a lot better than trusting a single consumer-grade harddisk to never fail. They all fail earlier or later.
does the desktop matter anymore? Posted Sep 28, 2005 2:18 UTC (Wed) by zblaxell (subscriber, #26385) [Link] why would anyone want to entrust all their private memories to a foreign company?
does the desktop matter anymore? Posted Sep 28, 2005 14:45 UTC (Wed) by mmarq (guest, #2332) [Link] Entrust data !
Google offered 2Gb for mail, they could easly offer as much for descriptionary user/client backup, now that single disk devices are at 1Terabyte,... no?
Is everybody so dumb they need photo applications do backups for them, or they can trust a simple dialogue a configure a automated backup for photos or other relevante folders, and many things more, from their desktops ?
In your line of operations, applications are king, and if consider many type of data, you soon get **LOCKED-IN** to particular applications and to perticular service providers, incompatible between eachother, preventing you from changing application... CONGRATULATION YOU ARE NOW A SLAVE IN M$ DREAM.
Me i rather TRUST MY DESKTOP, buy a external disk for autometed/manual backups, because the probability of 2 disks, the main and backup, failing at exactly the same time is very close to 0(zero)... and i only have to pay for the disk, say 200GB, that most certainly will be cheaper them to rent online space, from any angle you look at it.
So the Desktop matter more now then ever before, which wont invalidate me from renting online disk space or other valuable Web-Services that dont enslave me.
does the desktop matter anymore? Posted Sep 29, 2005 3:45 UTC (Thu) by zblaxell (subscriber, #26385) [Link] People do actually advertise outsourced backup services over the Internet. With "military grade" 56-bit DES encryption, even. If I could remember who they were, I'd check to see if they were still around.
The cost proposition for general backups is fairly bad, unless they are able to do something intelligent, like merge identical files from different customers together in their backup database. The bandwidth requirement can be quite high, you typically use a lot of bandwidth sending data that you don't necessarily retrieve, and in the (typical) disaster case, nothing short of 10GBE will be fast enough for an impatient and/or desparate user. This is all costs for the service provider that are incurred by users who hope they'll never really need the service. It's a hard sell to anyone who isn't already desparate to have an off-site copy of their data somewhere (and do you really want such people as your customers? What _is_ this data, anyway? etc).
That said, there are some advantages. If you back up your own machine, you'll need a gigabyte or two for all the system software and applications; however, if someone backs up ten computers like yours, then the marginal disk space used for the 10th customer is just their own personal files. If someone backs up her MP3 collection they downloaded from the internet, it takes up virtually no space as long as the same company also backs up the guy who she downloaded it from. I use this trick myself when backing up 30 machines at the office, and it saves dozens of gigabytes on tape. Over 10,000 customers, the efficiencies could be quite substantial...or the indexing overhead could be crippling. If you are running the service, and you know that the population of users is all e.g. Fedora 4 systems with OpenOffice, then you can give customers a discount on the contents of /usr (because it will be gigabytes of identical data, it will be cheap for you to store, and your customer has the convenience of not having to reinstall it all on restore).
I agree that application-specific web services (particularly those that are heavily optimized to be used by human browsers) are damned annoying things to use in combination with other application-specific web services. Sometimes it is possible to cobble together some Perl scripts that act as a gateway from one web application to another. Sometimes it is possible to cobble together some Perl scripts that combine a few desktop GUI components together, too, and increasingly a bunch of GUI components is starting to look and feel like a bunch of web servers (or is it the other way around?). Sometimes it is just easier to find someone who has put all this stuff together for you in some useful way.
does the desktop matter anymore? Posted Sep 28, 2005 15:57 UTC (Wed) by mmarq (guest, #2332) [Link] " Because so many systems are misconfigured, compromised, or just plain broken, such that they slowly corrupt data over time. A bit gets flipped here, a word gets doubled or zeroed there...unless you're verifying your data every day, it's possible to have this going on for years without causing anything to visibly fail. Since most people don't run 'md5sum -c' over their entire filesystem every day, most people don't notice the damage until it's too late. I look for it proactively, and find it disturbingly often. "
The big fault here, for not saying the only is of the concept of FILE SYSTEM.
I'm the one that belives that an enourmity of resources and money have been spent in the last 25 years building insecure systems... many other reasons came to mind besides technical insufficience.
And what is a persistente sinle level store that cuts down *for good* with all those nasty inconsistente file systems, without being incompatible with them.
does the desktop matter anymore? Posted Sep 27, 2005 13:44 UTC (Tue) by TwoTimeGrime (guest, #11688) [Link] > i see new desktops such as vista and osx etc touting operations such as> photo management etc...hello? don't people use tools like flickr for this?
I had never heard of Flickr so I had to look it up. It looks like a web site that stores your photos. I can't see that being useful for everyone. I don't see the appeal of people wanting to upload all of their photos to someone else's computer and have them store them. What if they lose all of your photos? I also don't want to share my photos except with possibly a handful of people (some friends, my parents).
> are there people who really only store their photos on their local box?
Yes. With over six billion people in the world you can't reasonabily expect that every one of them would want to do things the same way you do.
does the desktop matter anymore? Posted Sep 27, 2005 14:40 UTC (Tue) by mmarq (guest, #2332) [Link] " seems more and more, functionality is going straight through the browser stack (http->x/html->css/js/dom->user) and bypassing the desktop app layer entirely "
But dosent KDE and GNOME have a browser stack as part of their frameworks ?.
I belive what you mean is that we can stick that browser stack on top of a window manager and shed all the heavy management frameworks of the tipical desktop. But "thin or fat" you can't simply discard that desktop management layer unless you want to go back to lauch every application from a command line.
does the desktop matter anymore? Posted Sep 27, 2005 14:58 UTC (Tue) by man_ls (subscriber, #15091) [Link] Sure you can have each application implement their own, but wouldn't that be a terrible mistake, waste and fragmentation ?That is a strange way to think. Why is it a waste? I would rather say "a clear sign of a healthy environment". Perhaps if you are a company and could dictate what everyone works in, and you were the one paying for it, it would look like a waste. As it is, having at least two competing browser engines is great. I regularly use both Firefox and Konqueror; when one fails on a page, the other usually works fine.
does the desktop matter anymore? Posted Sep 27, 2005 15:57 UTC (Tue) by mmarq (guest, #2332) [Link] " That is a strange way to think. Why is it a waste? I would rather say "a clear sign of a healthy environment". "
A healty environment is one that *separates the application from the management of the desktop* in my way of thinking. Because that separation means that applications now could use descriptionarly(could be added/removed as plugins) all or any of the management features available.
Wouldn't it be nicer if instead of changing from Konqueror to Firefox it was possible to change from khtml to geecko, and that konqueror could have all tose extensions from Firefox *and more*, transparentely applying to it the same way they could apply to Firefox ??.
This means that applications are basicly made of *COMMON* smaller building blocks, and not a big solo monolithic block, and even ISVs with big, fat, feature bloated apps could roll new versions easly and in short time by only rearranging the blocks and adding a few others.
It dosent have to be like trying to tie different components togheter has so far has been, which as proved to be very clumsy, but it could be something like XUL tying the visual aspects of toolkits, and a common engine like " RuDI " to make all other different components of toolkits and or applications talk to eachother!
Well at lest RuDI is on the right path... i vote it for a sure sucessful Freedesktop standard.
does the desktop matter anymore? Posted Sep 28, 2005 6:52 UTC (Wed) by man_ls (subscriber, #15091) [Link] So you also think that having two browser engines (in order to exchange them underneath the browser) is good? Then I don't understand anything.The idea of universal plugins has been tried, it does not work. Universal extendability means universal confusion and security holes, so that things tend to stop working. You acknowledge such a thing yourself, and then go on to speak about vague toolkits. Sorry, but it still does not work. I do not desire to have interchangeable browser engines. I don't want to separate application from management. I just want good, robust desktop programs; if I can choose between two or more good ones, even better.
does the desktop matter anymore? Posted Sep 28, 2005 15:09 UTC (Wed) by mmarq (guest, #2332) [Link] " I do not desire to have interchangeable browser engines. I don't want to separate application from management. I just want good, robust desktop programs; if I can choose between two or more good ones, even better. "
With respect but you are confused !?... because what you express is exactly what you have now. And if your robust big applications start sharing common elements most probabily you wont notice a thing rather then everything having a more common look/behavior in dialogues and other parts.
Just read about RuDI http://www.kdedevelopers.org/node/1398 , is completely orthogonal to applications, **you can always have your big monolithic app** if you wish.
I dont know about interchangeable browser engines, but i belive you gonna start to have much much more separation of applications from management, simply because big monolithic apps are very expensive to develop and worst than greesy pigs to maintain.
does the desktop matter anymore? Posted Sep 28, 2005 15:51 UTC (Wed) by man_ls (subscriber, #15091) [Link] With respect but you are confused !?... because what you express is exactly what you have now.With respect too, if we really had good robust programs people would not complain about desktops. **you can always have your big monolithic app** if you wish.As a matter of fact, I do not like big monolithic apps (the proverbial "swiss knife" paradigm); I rather prefer small lean tools (the "set of knives" paradigm, you might say). Even better when engine/daemon and GUI are separated. From what little I understand, the RuDI framework you are referring to is a different concept; some kind of "protocol" to "interoperate". The mention of SOA (Service Oriented Architecture), BPEL and WSDL near the end scares me a lot. I have had to suffer two of them at work, and they are horrible, hairy beasts. I dont know about interchangeable browser engines, but i belive you gonna start to have much much more separation of applications from management [...]I thought "interachangeable browser engines" was what you meant when you were talking about using Konqueror and changing from Gecko to KHTML dynamically, sorry if I misunderstood. I still hope that more people get the Unix and X way to develop programs. IMHO client/server is a great paradigm and has a lot of life left, when implemented right. I don't see the need for these "advanced" concepts which just reinvent the wheel every few years.
does the desktop matter anymore? Posted Sep 28, 2005 19:51 UTC (Wed) by mmarq (guest, #2332) [Link] " if we really had good robust programs people would not complain about desktops. "
But we have good and robust programs, some of the very best in all IT world. People complain mostly is about desktops because they want consistent *good* look & behavior. They want the management of the desktops to be more intuitive, interactive, and above all be consistent with all the programs/applications.
Better or worst they will soon seam to be wanting to have a *butler* inside their destops, being them at a thin client in a completely server/client approach or a fat PC, that responds to them with 'yes master' and above all do the *cleaning* of the house for them automaticly and transparentely.
Desktop is a completely different methafor, which main function is to present/expose and *manage* interfaces of the systems to the users.
They (programs and desktops) should cooperate heavly and not be incompatibles. So If all applications could use desktop services or features for better interacting, where could it be a problem or a clash ?
" I thought "interachangeable browser engines" was what you meant... "
That is just a *possible* (and i say possible because it is so) example of heavy interaction between applications/programs taking advantage of desktop services/features.
" I still hope that more people get the Unix and X way to develop programs. IMHO client/server is a great paradigm... "
Agreed. Striking example is Linux Terminal Service Project(LTSP) serving dozens of *Graphical Desktops* with dozens of applications/programs running(even other services/servers) to diskless old or cheap worstations, and all that from a single server. So if i understood correctly what you meant, the traditional Unix and X way is not dead, it only evolved and changed to a better form.
" I don't see the need for these "advanced" concepts which just reinvent the wheel every few years. "
That is your prerogative, but belive me that nothing stays exactly the same, not even for a picosecond.
My prerogative is that i always did and still do complaint about the lack of firm and decisive move into the desktop arena of Linux/Open Source, forcing the evolvement and changing of the old Unix and X way in the process. My view is that the real power in IT as in many other fields resides on the masses of users and not on servitude(servers/services). The goal is for Linux/Open Source to gain more traction and last longer, because i have no ilusions and some day Linux/Open Source will be replaced or have to evolve and change to something different and more fit to future problems and paradigmas.
|
|||||||||||||