Moving to the Linux Business Desktop
M. Gagné, a writer for The Linux Journal, does not assume you're going to use any specific distro for Linux. He gives instructions and examples for the most common ones: Fedora (Red Hat), Mandrake, SUSE, Debian, etc. KDE is the primary desktop, but GNOME is covered fairly well, too. I have to admit that, as a long-time Red Hat user, I was well entrenched in the GNOME world. However, after reading Marcel's book, I've make KDE my default environment, and I've been very happy with it.
This book is broken up into three major parts: Getting to Know Linux, Administration and Deployment, and The Linux Business Desktop. Each part is packed with information in an easy-to-follow format. In fact, I found it hard to just read and not fire up my Linux to follow along.
Part One (Getting to Know Linux) covers the essentials of installing Linux and customizing your desktop. As I remarked earlier, Marcel covers multiple distros. He includes instructions on how to install using Mandrake, Fedora Core 1, and SUSE. For those of you who just can't wipe Windows from your hard drive completely, M. Gagné covers setting up a dual-boot environment clearly enough that you will be able to have the best of both worlds.
The second part (Administration and Deployment) assists in setting up a fully functional business environment. In Chapter 7 (Installing New Applications), Marcel covers the various installation programs available across the distros. SUSE's YaST2 installer, Mandrake's urpmi, Kpackage (from the K Desktop Environment), rpm (the shell program), dpkg (Debian's package manager) and apt-get are all covered. In addition, he gives a clearly written explanation of how to build from source (The Extract and Build Five-Step -- page 124) that dispels any anxiety a newbie to Linux might have.
The next chapter covers the device support in Linux. When I started using Linux, device support was spotty at best. Now it's tremendously improved. Marcel shows you the basic of Linux's support. He then goes on to explain about network and Internet connections. Unfortunately, there is one major piece of errata in this area of the book. During his explanation of the difference between Class A, B, and C IP addresses, the information for class A was inadvertantly switched with the class C info. I've been informed that the errata is corrected on his website (www.marcelgagne.com) and in future editions of the book. Outside of that one unfortunate error, the rest of the book is pretty clean.
Later chapters dig into the topics of Backup and Restore (the most important and most underutilized functions), printing, email, web servers, file sharing (both Windows-like with Samba and Unix-like with NFS), thin clients (server-side and client-side) and desktop remote control. He even includes a chapter on installing and configuring LDAP (something rarely written about, but becoming more and more important).
The third and final part of the book covers the usual business applications. Email, arguably the "killer app" for office environments, is addressed first. Focusing on KDE, Kmail gets the lion's share of the coverage, with Evolution following behind. Desktop organizers come next, with Korganizer the favorite and Evolution (again!) nipping at Korganizer's heels.
The web-browsing chapter focuses on Konquerer, KDE's jack-of-all-trades application, and Mozilla. Most notably, significant coverage is given in the next three chapters to OpenOffice and its basic applications Writer, Calc, and Impress. For working with images, digital cameras and USB scanners are covered, with The GIMP as the preferred image editor. On-demand contact via instant messaging and video conferencing rounds out this marvelous book. Kopete and GAIM are discussed in depth for the IM arena, and GnomeMeeting for the VC work.
As with most Linux books, a CD is supplied. However, this book does not give you a specific distro for installation. Instead, Marcel chose to include a branded copy of Knoppix, the CD-bootable Linux. The idea is to let you play around with the various aspects of Linux using Knoppix before committing yourself to the actual installation.
All in all, this is a valuable book, covering most of the areas a business user wants to address. Notably lacking was coverage on how to try to run Windows applications under Linux. At the top of the review, I mentioned I keep trying to steer away from Windows as much as I can. Unfortunately, I usually have a couple of applications that I need but don't come in a Linux version. Even though VMWare, Win4Lin, and Wine were mentioned briefly, I would have liked to have read some examples of running a Windows application using them. In addition, the major snafu with the IP address space marred an otherwise excellent book.
You can purchase Moving To the Linux Business Desktop from bn.com. Slashdot welcomes readers' book reviews -- to see your own review here, read the book review guidelines, then visit the submission page.
Yes, but not everyone has everything they need right there. In my company it is almost impossible since we have windows only software that we can't port, or use wine for (believe me I've tried) This is for the banking industry. So until there is a way to make that work, I've done what most places have done, put linux in the server room.
Sure, Linux can work fine as a business desktop for those who want to use it as such. What about the working stiff's in the accounting / secretarial pool that could care less, know enough Windows 2K/XP to get the job done and would need a 2 week special high intensity training course for dummies to learn where all their new tools are? These are people who would rather be fishing or watching the soaps, secretly despise having to work at all in an office, dream of winning the lottery, and resist change or having to learn something different, worry about being able to transfer these skills to other offices that are likely Windows based, etc.
Just playing Bill's advocate here.
try { do() || do_not(); } catch (JediException err) { yoda(err); }
If you need a book to do it, the gap has not been filled.
-- People who hate Windows use Linux. People who love UNIX use BSD.
The people you call working stiffs certainly wouldn't need a 2 week special high intensity training, that's just ridiculous.
You make it sound as if a secretary typing letters all day in MS Word would need to go through a boot camp from hell in order to be able to do the same in Writer and that is simply laughable.
And you forget the most important advantage -
When stupid users open that latest Funny.exe file, nothing happens! For that one reason alone, I think a Linux destktop would rock.
But the advantage of Windows is more psychological and social - there are jobs where if you put, 5 experience working in MS Excel would get you the job - however, people would not know what OpenOffice is at all. So, from that point of view, people may not really like switching over. It's got to be a gradual process, where they are first acquainted with the fact that an alternative exists, and then move on.
At home, where there isn't a system administrator to take responsibility for everything, something like OSX might make more sense for some people. For a business large enough to have that fulltime system administrator, it seems hard to justify not going with Linux.
See what I've been reading.
Yea, unfortunately we can't afford to buy 200 copies of vmware, plus the os licenses, plus the fact half the people would be clueless as to what they were doing. Maybe one of these days....
Come to think of it, I believe the problem is rooted in two fundamental beliefs of the open-source world. Number one: "Release early, release often" -- personally, I prefer to focus on productivity, rather than on backward compatibility issues. Number two: "Don't tell us what to develop, or how to develop it" -- sure, but if you don't develop software that addresses unmet needs of the business world then business will look elsewhere.
You seem to have confused "laser printer" with "cheap inkjet printer". I've never had a problem with a laser printer not working on Linux.
Any laser printer that doesn't support a sensible set of printing protocols (postscript, for example) does not belong in an office. It's fairly hard to find one, but if you're looking for something that will be absolutely no use to anyone, look no further than the Epson Acculaser C900. One of our clients bought one of these for their accounts office, where they have a high printload. Once they realise how much it would cost them, they sent it into the MD's office (which does very little) and replaced it with one of these. Since then, the Epson has broken down twice.
Kindly recommend to your boss that any money saved by buying cheap GDI printers is lost very quickly in maintenance and consumables.
"The dew has clearly fallen with a particularly sickening thud this morning"
The problem with using Linux when the people you work for generally use Windows is, of course, being compatible with them.
Odd that really. If you have a mixed shop with Linux, MacOS X, Solaris, and *BSD everything plays very nicely together. It is very much Windows that is the odd one out here - very much Windows that doesn't play nice with everyone else. That means that should Windows actually lose some real market share and not, by default, be the absolute dominant force that everyone else is forced to be compatible with... well, all of a sudden that lack of playing nice is going to look very bad for Microsoft. It's all about mindshare. Right now MS has it, but a little slip can cause a dramatci change.
Jedidiah.
Craft Beer Programming T-shirts
I'd be careful where you use that word. I'm an advocate of OOffice, but it does have its downfalls.
:(.
Open Office does not read word documents flawlessly. That I can attest to, for sure. Where I work, we discussed the possibility of switching over to open office, but the reasoning behind getting skrewed out of even more money from MS (alot more), was because ooffice did not convert doc and xls files correctly.
This wonderful suite is very unfortunately, not compatible enough to be used in a corporate situation
Codito, ergo sum.
Actually no, GNOME is not offering any real live applications for science, industry and students. Sure you are permanently keep refering to Evolution here but Evolution is a PIM. One of many tools that people require. A Secretary or Project leader might have a huge use of Evolution like managing their contacts, schedules and so on. But an IT professional who has to do UML based diagrams for his customers has no use of Evolution. Another one has to do presentations using a Presentation app like KPresenter even he has no use of Evolution.
I am talking about true life scenarios here and not programs that an ordinary homeuser would use. GAIM, XChat, GnomeMeeting are nice and KDE offers alternatives as well but they are not really practical in a money making business. They are only usefull for communication but not for production that get's them money.
There are plenty of GNOME applications but none of them of high scientific quality that I would like to recommend people. I do come from the GNOME camp and spent a couple of years with GNOME so I know what I am talking here. It's not that GNOME is unknown to me but also knowing about professional development I can say that GNOME lacks a good architecture to develop applications.
GIMP and Mozilla are by the way pure GTK+ applications they have nothing in common with the Desktop GNOME (which I was refering here). Mozilla also exists for KDE using KDE libraries but this doesn't really count.
XChat, Gaim, GnomeMeeting, Rhythmbox, Totem, Evolution and so on are nice but not what people within companies, research, science and information technology is requiring.
From your reply I have the feeling that you must be quite young (maybe beginning 20 or 22) I would urge you to visit 8 semesters computer or computer and economics science at an university. Before I went to university I had similar ideas like you but that got changed once confronted with real life and real life applications and scenarios. Stuff that gets you money in the pockets and right now KDE is leading here (from Open Source perspectives) of course real scenario companies still prefer Windows.
Let's not forget, te focus here is "business desktops". Not "development desktops". That means we need 100% interoperability with a variety of MS document formats, including:
- Word
- Excel
- PowerPoint
- Project
Word and Excel are mostly there, but PPT is iffy, and I'm not aware of anything for the OSS desktop that is 100% (or even close to that) interoperable with Ms. Project. If someone can point me to solutoins to those two problems (PowerPoint, Project), especially if there are free or reasonably priced, well supported versions for both Linux and OSX, we'd be down to 3 WIndows users within a week (from 10-12).
that ldap would be central to this book. How are you going to manage user accounts with linux desktops without it? One could still use NIS (which is easier), but that doesn't play too nice with windows. With samba/ldap/linux combo, you can truly have a multi protocol auth server with everything stored in a directory. What does the author reccomend as an authentication system?
the main issues to me with linux desktops are:
* authentication system (needs to be cross platform), meaning pam and ldap
* automounter (for roving home dirs, etc)
* nfs
You says everything was "server oriented" but that's how it should be - if your linux desktop isn't centrally managed you're doing it wrong.
I think Marcel's books are inspiring and I buy and read them, and act on them. Recommended!
Having said that, my company is a good example of Marcel's target. We are small (100 people in 4 countries) and techie (we have competent and motivated Linux techs, managed by me, a CTO who likes Linux). And yet we have not rolled out large numbers of Linux desktops.
Why not?
1 - User resistance. Cries and shouts from users and "We do not have time for that now" from techs. I think this is a simple one to overcome and that is my task - management needed.
2 - Apps. Our accountants use Quickbooks. Graphics guys use Photoshop. And so on. This is the real killer.
The OS is solid, Security is great - better than Windows. The only problem is that while 90% of the apps are fine - OpenOffice is perfect; media players can be installed and they work - the remaining 10% are showstoppers for 80% of the people.
Take me as a typical business example. Look at my laptop. Follow me from A to Z: My apps are:
- Various Canon digital photo apps for my 20D camera. Digial Photo Professional and the CR2 reader. No alternative: I need a Windows PC.
- CorelDraw - I guess I could find an OSS alternative... not as good but just about doable.
- iPod software: perhaps there are OSS alternatives but if so I doubt they are very good, and in any case they will need much time to get them working.
- Mozilla: OK in LInux too
- OpenOffice: same!
- Nero: alternatives available
- PGP: same
- Photoshop: no alternative at all. Photoshop is not available under Linux and nothing else comes close in the photography world.
- Quicktime: I imagine I can read Quicktime files in Linux, probably; no big deal anyway really.
- Ixdirect CRM: can run under Wine if we put our minds to it.
- MSN messenger: alternatives and clients available in Linux.
- Realplayer: can I play Real media in Linux? No idea but I imagine perhaps so?
- Outlook Express; no problem.
So, Photoshop (please do not suggest Gimp comes even remotely close!) and the Canon software and maybe the iPod software - that is all - but all that is a real showstopper. As long as there is no Photoshop for Linux I will not move my laptop.
And 80% of my company have some such killer app that runs only on Linux.
That's where we are. If the US court had shown some balls and forced MS to spilt OS from apps, by now we would have had Office for Linux and hence also all the other apps for Linux. Since they had no such balls, we will be in this limbo-land for years to come. Pity.
I wil get on and move the 20% (e.,g. helpdesk staff, shipping staff), anyway...
Michael
---
BDOS ERR ON A:>