sellis

That's an interesting idea. I tend to run MDI programs full-screen anyway, so each one is a bit like a mini-desktop anyway. Using multiple desktops should be a pretty good improvement. Now, how to re-enable multiple desktop support... I don't want it above, I want that screen space interdicted for all windows, above, underneath or newly created. Look at the way docking windows work in, say, Visual Studio. My project pane is always on the left, and cannot be overlayed or underlayed by anything. It's not "on top" of the document windows, you simply cannot move windows into that space. THAT's the functionality I'm missing. I hate floating tool windows. To paraphrase Alexei Sayle, I hate floating tool windows more than I hate fascism. Paint Shop Pro's tool options window is a floater now, and I'm constantly having to move the bastard thing out of the way. GIMP has this problem in spades. Same here, but now, instead of just pressing a key to tab between panes, I now have to mouse down, choose from a list of similarly named windows (because they all have the app name in them too) from a pop-up list, and go on from there. Poor substitute, IMO. I do it quite a lot. If I decide to change the name of a method, it's defined in one file and used in 5 others. I want to use the same search terms across all open documents in the application. Verrry friendly... ;-) What about new windows? In MDI apps, they have a well-defined area they are allowed to inhabit. Your toolbars, etc. are in consistent places and are never overdrawn by existing or new windows. As I say - toolbox vs. sack. Not necessarily. The way I browse through books on a bookshelf, versus the way I thumb through the pages of a book, are very different from each other. I like to manage related tasks like the pages of a book (bound together, distinct from other books) rather than as a pile of loose leaves. But, thanks for the tip about desktops. That goes a long way toward what I wanted. Edit: Fixed formatting tags - for some reason they all came out as text.

I'm in two minds about this. It could go either way. The good way: Linux is being taken seriously by M$, which means that lots of people may give it a look-in where it wasn't available before. This may give it traction in a business marketplace. The bad way: support is flaky and problems are blamed on Linux. Also, people considering moving away from M$ to Linux may just take this support as an excuse not to bother migrating.

I'm grudgingly getting used to the SDI-only thing, but I still don't like it. If it's the window manager's job, then it's failing. I do some development work at home, and am constantly frustrated by things. It's a bit like having a nicely organized toolbox at work, but a only a sack at home. I can't organize things the way I want and expect them to stay there. Imagine I have 12 Java files open (using KEdit), and XMMS running, a couple of terminals open, and a couple of images being edited in GIMP. (That's more-or-less the state of my desktop at about 10pm yesterday.) How do I: - Make sense of the toolbar now it has 18 windows on it? - Make sure that none of the GIMP windows ever overlap the tool palette? - Switch quickly between the Java windows only? - Search and replace for the same expression in several Java windows? - Get the same layout of windows (and constraints on their positions) next time I open the program? These are all things I can do with Programmer's Notepad and Paint Shop Pro on Windows with no problem at all. Anyway, just the opposing view. MDI is good. Whether this is managed by the WM or the app, I don't care.

Had another reply from a MEP today, this time Labour's Peter Skinner. No word yet from the Lib Dems, though, which is disappointing. Oh, and I've OCR'd all the replies now, too, for all you grep-junkies out there... web page still in the same place.

Not completely the same, but I wrote about my experiences with a 1215 scanner/printer combo here: http://mandrakeusers.org/index.php?showtop...67entry171967 This may give you some pointers...

Ahh. In the emacs-vs-vi holy war, the only winning move is not to play. I use kwrite almost exclusively - in addition to folding, syntax highlighting, being configurable enough for me, etc., it has those new innovations "the mouse pointer" and "the menu bar". The vi vs emacs holy war is as pointless as arguing whether the Ford Edsel or the Chevrolet Corvair is the better car to buy this year. Of course, now I'll get both sides of that debate flaming me. Oh well...

I recently installed version 2.0, and it's a big improvement over 1.x. However, I am still a fan of the docking windows/mdi paradigm. It allows me to organize my work area consistenly for that program, so that new windows don't overlap my tool pallettes, and so forth. I'm not a heavy user of GIMP, so I expect someone to pop up at any moment and say "Wait! Of course you can do that!".

Here's a great link for anyone fed up with the GIMP's UI: http://plasticbugs.com/index.php?p=241 Steve Moschella, of Plastic Bugs, has taken the core GIMP and reworked the UI so that it looks much more like a conventional application (in this case, Photoshop). A RPM is available too at codemills.com, although I haven't had a chance to download and test it on MDK 10.0 yet. Edit: On the codemills blog, Brent Shellenberg says "The RPM installed just fine in Mandrake 10.1 (powerpack). No dependency issues at all. Cool." But remember, your mileage may vary... Edit 2: The RPM has been taken down due to bandwidth issues. It is apparently available on bitTorrent though.

On the slow boots, is your network running? Next time it happens, type /sbin/ifconfig into a shell and take a look at the results. I had a similar problem with performance if the DHCP lease wasn't renewed correctly; it seems that a lot of programs want to know the computer's own IP address and if it doesn't have one then it takes a while to time out. On a reboot, it may then correctly reinitialize the network subsystem and work from then on (for a while) until the DHCP lease expires again. If you're on a home network behind a router, try setting up a static IP address - that's what I did.

Yes. The port with the "DV" and the stylised "i" logo next to it is the FireWire (iLink) port. You will need to use this to download the video with dvgrab. Of course, you will need the right cable, and a FireWire port on your computer, too.

USB? I think you have to connect via FireWire (a.k.a IEEE1394, a.k.a. iLink). I just downloaded some video today from my Panasonic cam, using a FireWire cable and dvgrab (part of the standard 10.0 distro, if I recall correctly).

In the past few days, I've had 3 replies from MEPs: Conservatives (a bit noncommittal, wait-and-see), Greens (anti) and UKIP (very short and anti). I've updated my web page to include scans of the letters, so you can make up your own minds about what they have to say.

Take a look at this thread: http://mandrakeusers.org/index.php?showtopic=14972. It has a lot of info about getting ndiswrapper and all its extra bits and bobs working properly. It helped me no end.

FlexBeta did a comparison. Nero doesn't come out well. http://www.flexbeta.net/main/articles.php?action=show&id=86 Favourite quote: "NeroLinux is just not worth the time to learn how to use or even download while K3b is still around." (Edit - sorry: pasted the wrong link. The link above should now be correct.)

I had a similar problem, but instead of moving the HDD, I was installing a new motherboard. Here's the post summarising my experience with that: http://mandrakeusers.org/index.php?showtop...ndpost&p=168760

OK, I'll open this up into a proper religious debate... The GIMP's user interface is unutterably dreadful. No, not because it's inherently awful - if you did nothing but use the GIMP all day, you'd adjust quickly and you'd be fine. But the trouble is that everything is different from every other applicaion, for no good reason that I can see. I use it occasionally, and although I don't expect to remember exactly where the "cubism" filter is, I certainly shouldn't have to re-learn where the "save" command is every time. On all other programs I use, it's in File > Save or Save as... I can access it via the keyboard as Alt-F S or Alt-F A. On the GIMP, it's in the right-click menu, and this confuses me every time. Again, you may say, "you will get used to it". My reply is "Why should I have to?" Jacob Nielsen, web usability guru, sums it nicely in this page: http://www.useit.com/alertbox/991114.html. He's talking about webpages, but the following quote applies to applications too: The GIMP is a superb program, but in my opinion its power is much less accessible because its user interface sucks. Right-o, let the flamewar commence...

http://www.tldp.org/LDP/abs/html/commandsub.html Hell and bat's do's... that's amazing. I keep having to remind myself that bash scripts are NOT just windows batch files on steroids, but proper programs in their own right. Looks like the "test=$(<backslash.txt)" syntax will answer the original question, then.

I'm not sure what you're doing, but could you use sed to replace '\' with '\\' in the input file? This should read in as a '\' and you wouldn't then need a second replace at the end. This, of course, in case you can't actually suppress '\' expansion on reading. Also, have you considered using one of the specialised file manipulation languages (awk, perl) instead of a bash script?

Sorry if this sounds like a stupid question, but do you actually have it plugged in? I run my networking on wireless (wlan0) and therefore have nothing plugged into the ethernet port on the motherboard (eth0), and get this error too. I could just disable eth0, but I'm convinced that 2 months after I do, I'll need it for something and won't be able to remember how to turn it back on again...

Well, I went for the scattershot approach. Find out all the MEPs for my region, and write to them all. Good old physical letters, too, with real stamps - less likely to be ignored than e-mail. It's worth the couple of quid and a couple of hours to help defeat this, even if it does mean asking for help from the Tories... :-) Edit - Now also uploaded to my website, for the delight of the 7 people who actually ever visit it... http://homepage.ntlworld.com/seanellis/

Excellent news! I'm glad you got it working in the end. No problem! Just passing on the recommendation - I asked a similar question about how to connect using VNC to a machine behind a NAT router, over on this thread: http://mandrakeusers.org/index.php?showtop...63entry168163 So we should all really thank "papaschtroumpf" for the original suggestion.

Yes. I use it to ssh into my home machine from work. It's easiest to do this from the machine you want to create a name for, as the dyndns site will auto-detect the IP address. Just follow "Dynamic DNS" link from the home page (at http://www.dyndns.org) and follow the instructions. It should take about 5 minutes. You will need to sign up for an account, and then add your machine to your host list. It really is that simple. That's the most difficult bit dealt with, then. You don't have to have any additional software running on your machine, and the router will automatically inform dyndns after renewing the DHCP lease from your ISP. Your router should have instructions on how to do this; you will probably need to enable dynamic dns and enter your host name and account information.

Qchem, I just tried the ipchicken thing, and there's a problem. If you're running behind a web proxy (my ISP has one), then it's the proxy's address that it returned to you, not your router's IP. This may not be a problem for mikaowx, but it's a flaw in the plan. Anyway, have I mentioned dyndns enough yet? I'm convinced that this is the correct solution, especially when re-reading one of the earlier posts: This is exactly the kind of problem that dyndns has solved. No need for e-mail notification, your machine is always "yourhostname.dyndns.org", for ftp, ssh, http, cups, whatever service you're running on it.

No, they don't. They already have to have DNS configured to access the web server, and that's enough for dyndns to work. When you register, your router will get a name something like this: yourhostname.dyndns.org ...and you can then access this via any other computer connected to the internet, or use that name in links from the web server. The regular DNS system will take care of address resolution. So, instead of uploading something every hour (which means that you could be offline for nearly an hour if your IP changes at 10:01), which would contain something like this: <a href="http://85.42.161.79/yourfile.html">Link</a> you just have a link in the html like this: <a href="http://yourhostname.dyndns.org/yourfile.html">Link</a> ...and let the dyndns update client worry about the fact that "yourhostname.dyndns.org" is IP address 85.42.161.79 today and 85.42.161.111 tomorrow. It's simple to set up - especially if the router does it for you - and it is certainly much easier and more scalable than having cron jobs updating web pages every hour. It also gives you external access to your box for a host of other services too. I use my account for ssh and sftp to help me transfer work to and from home. Some links: How to set up a dyndns account (you get up to 5 free, for non-commercial use) http://www.dyndns.org/services/dns/dyndns/howto.html Automatic update clients (run in the background) http://www.dyndns.org/support/clients/dyndns.html

Just had a quick look at my etc/fstab file, where I have a 128MB USB2.0 pendrive set up. Here's the entry, in case it's of any help: /dev/sda /mnt/usb vfat umask=0,iocharset=iso8859-15,codepage=850,noauto,user 0 0