Some advice

[Zyrtec]

I'm going to be writing reviews of several different Linux Distros, and I was wondering if using the LiveCDs that many distros make possible are good enough to do a general review of the OS without doing a full install? I was thinking of investing in a second hard drive to strictly use for testing, but I won't be able to do that for a bit.

Also, which ones should I do reviews of first? I was going to do Lucid Puppy first, since it is designed to run especially well as a LiveCD.

Thanks!

[Quake]

I think you should install every distros that you want to review. It will give you a better impression about how a user will live with the distro on a daily basis.

Plus, there's a performance issue. You won't know if the distro is "snappy" enough due the the USB bandwidth issue.

Think of the liveCD as a display unit. The speaker you hear in the department store may sound good there but how will it sound in a true situation? (your room)

That's my 2cent.

[BoneKracker]

Running a live cd is only a good way to test a live cd; it's not a good way to test a distro.

I think you should start with a review of LFS, then you would learn things that would more fully qualify you to write the other reviews.

Next would be Gentoo, FreeBSD, Slackware, Arch, Fedora, Debian, and OpenSUSE.

Also, you should use a distro for at least a month, for a broad variety of tasks, and participate in its community, before you write anything about it.

[cariboo]

I know most of the reviews of distro's you see on the net are installed in a vm, that way you can make screenshots of the install process.

[Zyrtec]

I know most of the reviews of distro's you see on the net are installed in a vm, that way you can make screenshots of the install process.

This is a great idea, and I'm glad you mentioned it.


What's really going on, is that I'm going to be writing articles for Associated Content, and I'd like to get the word out there for Linux and FOSS stuff. I also get paid small amounts for page views :P

[BoneKracker]

I know most of the reviews of distro's you see on the net are installed in a vm, that way you can make screenshots of the install process.
++

That's a good tip.

Also, a truly complete review of a distro requires trying multiple installation methods. Most real distros have live CDs, text-based installers offering advanced options, minimal installs, network-based installs, online build systems, etc. In some cases, not all of these are good.

[BoneKracker]

This is a great idea, and I'm glad you mentioned it.


What's really going on, is that I'm going to be writing articles for Associated Content, and I'd like to get the word out there for Linux and FOSS stuff. I also get paid small amounts for page views :P

Well, if your primary goal is page views, don't worry about first becoming an expert. Just write something both controversial and arguably incorrect, and you will get page views alright.

You'll get flamed so bad you'll need therapy and have to change your identity, but you'll get page views. :lol:

[Shining Arcanine]

Running a live cd is only a good way to test a live cd; it's not a good way to test a distro.

I think you should start with a review of LFS, then you would learn things that would more fully qualify you to write the other reviews.

Next would be Gentoo, FreeBSD, Slackware, Arch, Fedora, Debian, and OpenSUSE.

Also, you should use a distro for at least a month, for a broad variety of tasks, and participate in its community, before you write anything about it.

This would be a good way of approaching things, although it is definitely time consuming. I do not think things must be this strict, but I think using LiveCDs alone as the original poster wants is inherently flawed, especially since some distributions put a great deal of work into the installation process. Of the options you cited, think that LFS put the most work into it.

[Zyrtec]

Well, if your primary goal is page views, don't worry about first becoming an expert. Just write something both controversial incorrect and you will get page views.

You'll get flamed so bad you'll need therapy and have to change your identity, but you'll get page views. :lol:

Hahaha, more great advice! I'm not worried about the money, though. I just hope to maybe sway more people to try Linux, and support open source.

And I certainly don't consider myself an expert by any means. I'm probably intermediate at best, but I'm hoping to write these reviews as an "average Joe" and give a completely honest opinion from that kind of view.

[Zyrtec]

This would be a good way of approaching things, although it is definitely time consuming. I do not think things must be this strict, but I think using LiveCDs alone as the original poster wants is inherently flawed, especially since some distributions put a great deal of work into the installation process. Of the options you cited, think that LFS put the most work into it.

I never said I only wanted to use the LiveCDs. I was wondering if it would be alright to do for the time being before I can invest in a new hard drive to use. The Virtual Machine advice makes things somewhat different now, though.

[Shining Arcanine]

I never said I only wanted to use the LiveCDs. I was wondering if it would be alright to do for the time being before I can invest in a new hard drive to use. The Virtual Machine advice makes things somewhat different now, though.

From my experience with having used virtual machines to try out both Ubuntu Linux and Gentoo Linux, virtual machines do not do operating systems justice, especially when you are trying to use a desktop environment.

[Zyrtec]

Well if I can get like 20,000 - 30,000 views on these first couple of dumb articles (that don't pertain to Linux distros), I can get the hard drive a lot quicker! I get paid $1.25 per 1000 views on each article :P

[uRock]

You can do reviews on either process. LiveCDs are used as such by some people. I have Puppy and a few others that I use on other people's systems just so I can have the Linux experience everywhere I go. A review showing that a LiveCD doesn't work will cause the devs of the said OS to work a bit harder.

Doing reviews on the installed version can be hard unless you are going on about how it works. Most people will be changing the look, so that part is harder to grade.

[BoneKracker]

From my experience with having used virtual machines to try out both Ubuntu Linux and Gentoo Linux, virtual machines do not do operating systems justice, especially when you are trying to use a desktop environment.

Well, by "virtual machine" I doubt he meant, "grab the vmware image for debian and plop it on your machine". I think he meant, "install the operating system into a virtual machine" (the equivalent of creating your own image).

Although it is definitely the case that cannot evaluate performance if you have done it this way (or even hardware-related matters, depending on the type of virtualization you use).

Also, I wasn't really serious about doing LFS first. Without meaning to hurt any feelings, my point was that people who aren't experts shouldn't do reviews (unless they've made it exceedingly clear that their review is focused purely on something within their bailiwick -- like aesthetics or ease of use from a novice's perspective). If they want to do more, they must first make the effort to gain the basis for insight.

There are far too many people already blogging out the wrong orifice without first making the effort to be able to speak somewhat authoritatively (not that this applies to anyone here -- we're just talking about approaches, and I suggest an approach that includes learning).

I think to do reviews of linux distributions, one must have a feel for the inner workings (things like init systems, package management, kernel configuration, and hardening), the breadth of approaches in use, the evolutionary history, and the degree to which various parties are responsible innovations). That's why I suggested the mix of distros I did, in the order I did (although putting LFS in there was probably going a bit too far).

You can't understand "Linux" without understanding what portions of it are taken from UNIX and what portions are different. You can't really understand (and therefore evaluate) Ubuntu without knowing (for example) how much of it is Debian (and where Debian originated), the extent to and areas in which Debian and RedHat have cross-pollinated, or the origin of automated dependency analysis in package management systems and its role as an enable of desktop linux for the masses. You can't evaluate SUSE without understanding the significance of their automated build system. You can't evaluate Gentoo without understanding the significance of its source-based package management system (and it's roots on the BSD side). And so on. There is a lot more to a distro than what color scheme it has and how fast it boots up on your laptop.

[Zyrtec]

And I agree with much of what you say, BoneKracker. I really want this to be a learning experience for myself, but at the same time, I hope to sway some people to take the step towards Linux. Many people are too intimidated by the sounds of the things you mention, and a real in-depth review would do hardly any good for these sorts of people.

This isn't to say that I'm going to write reviews on the more "difficult to learn" distros that say things like "stay away from this operating system because it is super hard to learn, and you have to use commands to get everything to work!!!"

I think the best way to tackle this is to start with the more user-friendly distros (I'm treading lightly here; I don't want to cause any holy wars) to get interest from the general computer user that is used to Windows, and cover the lightweight distros meant for older systems. I would throw some of the others into the mix, so I can get interest from people all over the spectrum.

I don't know D:

[BoneKracker]

And I agree with much of what you say, BoneKracker. I really want this to be a learning experience for myself, but at the same time, I hope to sway some people to take the step towards Linux. Many people are too intimidated by the sounds of the things you mention, and a real in-depth review would do hardly any good for these sorts of people.

This isn't to say that I'm going to write reviews on the more "difficult to learn" distros that say things like "stay away from this operating system because it is super hard to learn, and you have to use commands to get everything to work!!!"

I think the best way to tackle this is to start with the more user-friendly distros (I'm treading lightly here; I don't want to cause any holy wars) to get interest from the general computer user that is used to Windows, and cover the lightweight distros meant for older systems. I would throw some of the others into the mix, so I can get interest from people all over the spectrum.

I don't know D:

I think you'll be fine as long as you make it clear that the review at this point is from your current perspective as a user with X amount of experience and that it's focused on certain aspects A, B, and C.

If you're going to write a series of articles, you might describe it as your "journey", and actually capitalize on the fact that you are going to learn more as you go. Given that, you may want to evolve them into a sort of comparison and contrast format, because you will have new things to say about the distros you reviewed before, that didn't occur to you at the time.

But if you really want some page hits, create articles with titles like "Is OpenSUSE Dying?", "Fedora: LabRat Linux", "Ubuntu: So Easy Cavemen Use It", and "Gentoo: Linux for OCD Victims". :P