July 2005

Since other people can do Part II’s of remarkable things and thanks to Mary for explaining a bunch of things I am interested in, I can do a part deux about what I see being some of the points and profiles of Linux on the desktop. This expands on some ideas in my first blog post.

Linux on the Desktop - a Myth or Magic or its Real

Asa Dotzler awhile ago posted his extrapolations on what makes Linux ready or not so ready for desktop use. Its an interesting read and the comments frame what people consider the strengths and weaknesses of same. In his recent Part 4 he covers some interesting terrain as well. I believe that the whole corrolary becomes more complex when one introduces variability to the constants. Lets just say the question changes from “is Linux ready for the desktop” to “Is Linux ready for a POS Enterprise Desktop”. Then we get brand new methods of measurement and we need to apply other standards. We simply cannot use the same use cases for a personal productivity desktop versus a very focused and defined desktop that may only have one component. As Asa points out,

How does Linux improve here? The challenge is to find those areas where it’s valuable to change and make the transition as easy as possible (through good documentation, intuitive or easily learned UI, etc.) and to find those areas where change doesn’t offer enough benefit and make those areas as comfortable as possible. It is not necessary to be a clone, but it’s foolish to deviate from what Regular People expect when the value of that deviation isn’t extremely high.

My answer is that Linux must improve on a series of paths instead of just one. There are different success ratios for each type of use we define and to make Linux work in each one there are different sets of success and failure criteria. I don’t know how many there are of these but I suspect that we each can define a few when we think of how Linux could potentially be used in corporate or enterprise or home systems. It becomes a “habilis” question or a tool user type thing. We need to retain flexibility as we seek out where it could and will work versus where it needs to get better and not where it needs to emulate Windows. People don’t want/need another windows clone. The diea is make something sufficiently different and by that difference people are attracted. Asa uses the term “Regular People” above; but I maintain there are no regular people in this game. Everyone has sets of requirements or tools or productivity requirements that Linux may or may not fulfill. A project manager may find that the desktop does not provide sets of applications to open visio or MS Project diagrams easily. Is this the desktop or a focused application? And where the two interact and integrate. The question must be posed where the destkop ends and the applications begins.

If we only ask for the so-called “Regular People” then we get “Regular Answers”. But those, I maintain, are not the answers that really count since everyone uses their systems in non-regular ways and have their own unique habilis requirements. I like Asa’s analysis but I think when we simplify things we don’t reach the core issues.

I’ve been studying the Linux landscape for quite awhile besides the job I now find myself in and loving. Often, one can find interesting points or views emerging with technology news these days. There has been press of late regarding the ability for Linux to assume a more significant role on the desktop and what it will take to get there. I’ve got a few impressions of this after reading the news that MS Vista probably will not launch on time (so whats new?). We tend to use the term “desktop” to have a certain meaning and we stack the term to include applications, utilities, integration and performance points that will allow a desktop user to become productive almost immediately. Note the qualifier in there being: almost. There is no way with current XP SP2 technology regarding updates and system management that a person can become instantly productive. The system is really not ready after an install and its actually dangerous to assume it is. Things are missing on a Windows desktop that are sorely needed. Consider the many types of nastiness that Windows does not protect you against with a factory default install:

• Anti-virus. You have to have this even if you don’t have SP2 becoming so annoying and telling you all the time you need it.
• Malware Protection. You don’t get any protection from spyware or other nastiness and default IE does little or nothing to help you out there.
• Registry and Performance Tools. Not existant. Go fish

All of these combine with the efforts at updating the core OS to a day or two of rebooting, installing, etc.

Lets move over to Linux and when we do many of these things just go away. We have no registry and I don’t know of any malware that really causes a problem to Linux. Don’t need antivirus protection really. But is a Linux system truly ready after an install? Or are there patches and updates there too? I would argue there are. You cannot just get done and have the latest applications and you need to run the system upgrade tool to get new stuff. This also takes awhile. If we move to the kernel, often on Debian I just run a net-installer but I am not done there. The kernel on the installer is butt ugly and needs replacing. You need a 2.6.recent kernel. But there is a more basic statement here. Its the term desktop. We affiliate desktop with productivity desktop I think and lump a bunch of things together.

What if I instead said corporate desktop or Point of Sale tool or a set of tools that have a set of requirements all carried out in a web browser? Then what happens? Well, this desktop differs from our personally run desktop in its needs. Updates and performance increases may not be so necessary and with Linux’s scalable installation approaches one can simply build their own installer. I think in arguments over whether Linux is ready for this or that, we tend to simplify the terms to a level that we easily understand. The truth I think is that there are different models of “desktop”. As we do technology consulting, we need to define carefully our interpretations of use and come away with sets of goals that represent the actual need. A enterprise or corporate desktop may not need all those nifty presentation tools. The world may be mozilla for them. Mozilla applications, interacting with databases in a web environment.

We always tend to simplify things so we can grok them better. Its our decidedly sociological bent. Technology needs to be divided into simpler components so we can grasp the requirements. But I think we need to be careful when we use encompassing terms like “desktop” and ask the question:

Is Linux ready for this or that? Is it ready for the desktop?

Then we need to apply what the unique conditions are. Its not just that Windows sucks rocks but when you get to a certain use case scenario there are the tools of the trade that make things work for you. Being a project manager, I have these sets of tools I need. I have to edit and update certain types of files with regularity. Is Linux ready for a Project Manager’s desktop needs? Now we can ask certain framing questions around all that and come away with the use cases and scenarios. In our urge to simply make a case for a thing, we want and need to limit the complexity to terms and conditions we understand.

The truth to me is that how a desktop is used and who uses it is far more complex than a simple answer can allow. If we read some of the recent blurbs of usability, integration, support for devices, etc. I have seen what exactly are those questions catering to? Define the type of desktop you want to support. I think there are so many different kinds of desktops that we ourselves and both the Windows and Linux communities a disservice by simply interpreting things at a simplistic level.

I’m forever amazed at the projects out there for consultants that may want to travel here and there, engage in meaningless Windows XP banter, install applications and hardware. I got this call from a headhunter asking me of my interest level in working over in Kuwait doing large scale IT infrastructure planning and delivery. Good grief! If I was a single stud, I may consider it. I can bet the IT infrastructure projects over there are simply huge. I read the IT Garage every so often also because it sometimes offers some view into IT and professional services. I’m still kinda interested in proserv but if I were given the choice of senior projject manageament or proserv management consulting, well… Easy choice. I was talking with my neighbor about the whole HP thing and what it may mean for jobs here. A few thousand more unemployed and we both think that many of those will simply leave the state and become another type of casualty. Are there IT jobs out there in other places? I bet there are. I think the silent IT enterprise hiring firms snap up certain candidates who pass their rigorous DNA, fingerprint, and coffee cake testing.

See?

Its getting closer to the bewitching hour so all possibility of rational dialogue is now ended.

I might as well blog what I really wanted to here. I have exciting news but I cannot remember it. Maybe it has to do with work. I don’t recall. Perhaps when I sit down tomorrow night, I’ll remember why I wanted to write tonite’s blog post. The Irish stuff though sounds uber kewl! There are my two required links to remain in my much vaunted position on the blogamid. The comet is twirling fellow blogmeisters! Hang on tight. Its gonna be a wild fucking ride tonite.