Print

Please login or register.
1 Hour 1 Day 1 Week 1 Month Forever
Login with username, password and session length

[vector]

Our very own Caitlyn has posted an in depth  review of standard 5.9. I thought it was quite good at pointing out the strong points as well as the weak ones and certainly things to ponder for the development team as well as the rest of the community. Here it is: http://www.oreillynet.com/linux/blog/2008/03/taking_a_good_long_look_at_vec.html
Wonder how the rest of you feel?
cheers,
Vec

[Lyn]

I thought it was extraordinarily good.  As you said pointing out the strong and weak points.  The question is do we want to go down the paths suggested, ie to be truly international and to be able to play with the big boys.  Do we have the expertise and the skills necessary to take the distro up a notch technically?  I know I don't.  Happy to try testing things but can't contribute much (anything) as a developer.    The bonus disk work has helped broaden what there is in the repository, which is a good thing.  That helps answer some of the issues with the breadth of applications on offer.  The question is where do we want to go today :-) A fully internationalised distro with options at install of installing in many languages - like Mandriva and fully localised  desktops etc.

[uelsk8s]

The question is do we want to go down the paths suggested, ie to be truly international and to be able to play with the big boys.

I dont know, I will leave this up to the community to decide. I do know that in IRC the other day we were able to help a new linux user add 2 additional languages to his system in less than 5 min, so I dont think we are too far off with that.

Do we have the expertise and the skills necessary to take the distro up a notch technically?  I know I don't.  Happy to try testing things but can't contribute much (anything) as a developer.

I hate when testers discount their Value to the distro. I think testers are one of the most important groups we have. Without testers we would have an OS that worked on only a handfull of systems.
We are sorely lacking a couple of things now.
1) package testers (we need people to download and do at least a minimal check on the packages in the testing repo)
2) Repo Maintainers (people that can move packages from testing to the proper repos)

Thanks,
Uelsk8s

[caitlyn]

First, thank you all for your kind words.  When I write a review I don't treat Vector any differently than any other distro I've worked with.  I try to be honest and as objective as is possible.  Everyone's views are colored by their opinions and I don't claim it's otherwise when I write something but I do try to be fair.

I should mention that I wrote a review of Ubuntu Gutsy Gibbon (64-bit) in January and then decided not to publish it.  Why?  It was intensely negative and harsh and I didn't want to deal with the reaction I'd get.  Let me just say that I'm still amazed that a distro could be released with the sort of bugs I encountered -- bugs which still haven't been fixed months later.

I agree with Uelsk8s.  Vector is very close and there isn't as big an effort needed as you might think to make it better than the so-called major distros.  I didn't say equal with; I said better.  It's mainly an issue of making a few areas more user friendly and adding some additional packages to the repos.  The installer also needs improvement.  By improvement I don't mean pretty or graphical.  Ubuntu only has a pretty installer on their live CD.  I mean things like getting video hardware detection to be as good as other distros.  I mean having an easy install path for newcomers.

I don't think the Vector developers lack any expertise.  That's been pretty well proven by the product that's out there already.

Regarding being truly international:  most major distros only support installation in a limited number of languages.   Ubuntu is in this category.  What they then offer are easy tools to work in any language you choose after installation.  Having said that the installer should have at least some multilingual capability for the most popular languages.  I realize some languages are incredibly difficult (Chinese, Japanese, Arabic) but others like French, Spanish, Portuguese, and even Russian don't require major changes in the structure of the installer -- just more choices.  We don't need to equal Mandriva or Fedora in terms of installer languages.

One thing Vector Linux needs and clearly doesn't have are a team of volunteer translators.  A good start would be to ask for translators right on the home page of the website.  Get the word out in the community that this is a concern to us.  The next problem once you have volunteers is quality control on those translations. 

Lyn:  Vector is closer than you might think in terms of internationalization and localization.  Set up a new user account with a default language and locale of French/Canada and see what your desktop looks like.  Most of the translation work is done upstream and it's already there.  Where it's lacking is in the unique to Vector tools and utilities and in applications from some of the smaller, less popular, often lightweight packages we include.  I don't think VL developers can worry about contributing translations to upstream packages.  I do think it is something to worry about for vasm, vasmCC, vwifi, etc...  In other words, it only should be a real concern when the farthest upstream you can get is right here.  I know that's a tall order. 

The other issue with internationalization and localization is ease of use. 

I dont know, I will leave this up to the community to decide. I do know that in IRC the other day we were able to help a new linux user add 2 additional languages to his system in less than 5 min, so I dont think we are too far off with that.

Yes and no.  I tried to help that gentleman in the forum.  He ended up having to edit his /etc/X11/xorg.conf file.  He's a bright guy who was willing to roll up his sleeves, get under the hood, and learn.  How many newcomers would feel comfortable doing that?  Not too many, I'm afraid. 

A number of us tried to answer questions for another newcomer.  He promptly abandoned us for Ubuntu just because we suggested enabling the testing repo and looking at linuxpackages.net.  Even worse, someone suggested building his own package.  Going beyond what gslapt showed for packages was just too hard for him to be bothered with.  He said he didn't have the time.  He wanted a distro that just worked and required no real effort at all.  He made the comment that he is typical of most users out there.  I hate to say it but I think he's right about that.  He won't be happy at all with 64-bit Ubuntu either but that's not our concern.

Do we need to please users like the one I just described?  Only to a point.  There are always going to be people who will balk at any learning curve.  Those are the folks who end up writing how awful and hard Linux is and how wonderful Windows is.  We can't worry about people like that.  What VL developers can do is make things as user friendly as possible so that a sane end user with reasonable expectations has a good experience.  VL has already come a huge way since 5.1 in that area.  This can be done without dumbing the distro down for experienced users, something VL developers clearly have recognized based on what's been done so far.

Things like being able to change keyboard layouts and language/locale from the GUI are really the next step as far as internationalization and localization are concerned.  That was promised for vasm in the forum a year ago.  The idea is to make it easy for anyone, even a newcomer, to use the tools that are already there.

Aagin, thank you all for the very kind words about my review.  I'll continue to volunteer and help as much as I can with packaging and testing.  The main limiting factor for me is time.  I'd love to do this full time and then some but that sadly won't pay the bills.

[GrannyGeek]

Thanks very much for your review, Caitlyn. It gave us lots to think about.

I'm thinking about whether VectorLinux should go for the maximum market (with the hope that Vec et al. can make a decent living through this) or for the maximum benefit to our niche. Not being one who would profit materially from VL's market success, I can't make that decision. However, from the purely personal standpoint, I'd rather see us go for maximum benefit to our niche. I like ease of use as well as the next person and very much appreciate the steady increase in user friendliness that VL has exhibited over the years. However, I also like that VL doesn't do *everything* for me. I like learning about *Linux*, not just how to use easy graphical configuration tools. To me, working with xorg.conf is one of the basics of learning Linux. So I think that any user who says "no way" without even giving it a try is a user who will be happier with a different distro. Editing xorg.conf is usually just copying something someone suggests. You don't have to know all the technical details. I'm willing to edit anything if someone gives me some guidance.

Good hardware detection is important, so I think VL should improve as much as possible in that regard (although hardware detection is another area that has improved greatly over the years).

As for multiple language support, I don't know what all is involved. If it's difficult, I think we need to be realistic. While VL should support common user languages, I think most users speak common Western languages. Supporting what are languages uncommon to VL users may not be such a good idea if it takes more effort and expertise than our user and developer community can offer. Besides, why not leave an opening for people in countries without proper language support in VL to develop their own distro? That's one of the beauties of Linux.

People who try VL and go to Ubuntu or some other supposedly easy distro because VL is "too hard" may be back in a few years. That was my history. I first started with Linux in 2000 or 2001, I think--maybe earlier. I tried VL for a few months but the learning curve was too much for me, so I went with Lindows (now Linspire) for a couple of years. It had a lot of ragged edges but it was easier and it kept me in Linux until I got tired of Lindows's limitations (limitations due to its dumbing down of "real" Linux). I then tried quite a few distros. My hardware was mediocre so my experiments were often disappointing. I then used SuSE briefly and found it very polished, but performance was not so great. I decided to try VL again and wow! The speed difference between Vector and SuSE and other distros I tried was amazing. That clinched it. Once I experienced the huge performance boost, I was determined to learn whatever I needed to know to use VL. That was about the time I think VL was making major advances in user friendliness. VL 4.2 was the version that made me settle on VL as my only distro. I haven't tried anything else except through Live CDs ever since.

You are a great asset to the VL community, Caitlyn. Whether or not I agree with your comments, I always benefit by reading them.
--GrannyGeek

[toothandnail]

One relatively simple area which I think Vector needs to fix is basic locale settings. There is a good range of locale data to be found in /usr/share/locale, but no effort is made to help users select the relevant data. Instead, /etc/profile.d/lang.sh is shipped with the assumption of en_US (not even en_US.utf8), and new users are given no guidance as to changing that setting. It affects a number of things in using Vector, and really does deserve to be set correctly during install.

As a simple example, I wasted a great deal of time trying to get Abiword to use the en_GB dictionaries, without any success. Eventually, reading through some data on the Abiword homepage, I discovered that it took its native language data from the LANG variable. After more investigation, I then found that I could edit /etc/profile.d/lang.sh to reflect the correct locale, at which point, Abiword and several other things started working as designed.

Now, anytime I install Vector, one of the first things I do is edit /etc/profile.d/lang.sh to put en_GB.utf8 into the LANG variable. But I really don't think I should need to do so in any modern Linux distribution.

paul.

[vector]

And we have a new review of our favorite vectorlinux-5.8-SOHO here:
http://calummegan.wordpress.com/2008/04/04/kde-built-for-speed-vector-linux-58-soho/
comments?
vec

[Lyn]

Very good, love to see a review of 5.9 back to back...

[Joe1962]

And we have a new review of our favorite vectorlinux-5.8-SOHO here:
http://calummegan.wordpress.com/2008/04/04/kde-built-for-speed-vector-linux-58-soho/
comments?
vec

That sounded suspiciously familiar, so after a couple of unsuccessful searches I reread most of this thread again and found it:
http://www.vectorlinux.com/forum2/index.php?topic=98.msg21952#msg21952
 

[Masta]

haha, ah sheeze ... looks like someone "trying" to be a blogger and copying someone elses work and using little modification. Oh well,, they tried,, lol.

[kazuya]

I used vector 5.9rc and was left with an uneasy feeling about VL. I am glad to say Vectorlinux 5.9 official is near flawless for me. I am surprised at the amount of improvement done on b/w the two.

I have installed kde over my install. All I can say is wow. It really is a shame that we do not have the kde desktop shown in its glory as we do post install. Most PCs are getting more fluff, but VL with kde still runs like a champ and has a more complete feel about it. Nothing against xfce, but the kde is simply stunning The combination of the themes, wallpaper, icon sets, speedy functionality, etc makes it a joy to use and work in.

I love the Nvidia or graphic card checker. it worked flawlessly for me on my HP machine. Prior to install, it warned about network card not being detected, but network ran great without me doing anything.

This distro continues to impress. I am yet to run an OS so far that parallels or surpasses the kde implementation in VL. Great work guys, keep it going.

I hope someday that gnome would be easy to include in VL along with e17 - but at least what is used in this distro is well done. I missed not being able to upgrade all, but at least when I upgarde an application, it upgrades the app and all its dependencies. That was a huge plus.

This OS is very robust - and Im thoroughly enjoying using it primarily now again alongside my other OS of choice, Mepis. lol.

EDIT: gnome via gsb installed and ran flawlessly. So now i have all the DEs i want with the exception of E17, and it still runs wonderfully. Great release guys.

[vector]

New Review of SOHO 5.9 can be found here: news.oreilly.com/2008/07/vector-linux-soho-59-deluxe-no-1.html
Caitlyn Martlin from oreilly.com did a bang up job and is a member of our forum!!

Cheers,
Vec

[dawnsboy]

This is a particularly well written review.  I appreciate the fact that Caitlyn has covered this release in depth.  Given the improvements in this release I may take another stab at installing VL on my Asus Eeepc. Previous releases work just fine (some manual configuration required) with the exception that the wifi card does not play well with VL. 

I have personally experienced the fact that with the release of VL 5.8 the distro has matured to the point that it can be used by newbies and novices alike when pre-installed in a new PC.  That is a critical point in my opinion.  Windows has the advantage of being onboard almost every PC on the market.  In order to use linux one usually must install it after the fact on a PC that was built with a windows os in mind.  This makes it essential that linux distros aimed at the broader markets must come equipped with user friendly installation scripts that automate the process as much as possible.  This includes the need for accurate detection and recognition of hardware that results in the installation of all required drivers during that process.

Linux distributions have a very large number of open source applications available for inclusion.  Many distributions succumb to the temptation of offering as many of these applications as can be crammed onto a dvd or multiple cd-set.  One of the first things I do as a user is strip out all of the applications that I will not be using that can be safely removed and then add the applications of my choice from the repository ( the place where most packages belong in my opinion).   I have the default Xandros OS on my Eeepc.  It occupies almost 2.5GB of the 3.7GB available on the SSD.  By the time I got done stripping out packages that are of no use to me and adding in the ones that I will actually use I had a net gain of over 1GB of drive space.  It is conceivable that I will never need to add another package during the time that I am actively using this little notebook.

Other users will have different requirements regarding the type of applications that they need or want.  Many of these personal requirements are met in VL by offering Light, Standard and SOHO editions.  I think that the most critical point to discuss is the "maturity" of the distribution as Caitlyn has so aptly pointed out.  All of the variations must share one thing in common; the ability to install easily with automated (as much as possible) hardware detection and installation.  The other point that seems important regards maintaining focus during the development process.  If a distribution seeks to be an enterprise solution then by all means equip it to do exactly that.  If a distribution is aimed at the individual user then equip that distribution accordingly.  Don't get carried away with including everything but the kitchen sink (oh wait, I forgot in linux you can include that as well  ).

Way to go VL.  It is made quite clear in Caitlyn's most recent review that the distribution is maturing nicely.  She does a wonderful job of highlighting the features and benefits of using VL while presenting a straight forward analysis of those areas of development that still need work.

[caitlyn]

Thank you both for your kind words.  They are appreciated.

I do try to be objective.  I remember worrying about what the VL community would think about my review of 5.8 Standard back in January of 2007.  It was, IMHO, a very mixed review.  The response was overwhelmingly positive and any criticism was taken as constructive.  That really impressed me.  A lot of distro communities are dominated by fans who attack anyone who says anything less than complimentary about any aspect of their favorite distro.  VL developers instead went about the task of fixing and improving things.  The net result is that each release is better than the last as there isn't that much left to criticize.  I've noticed that most other reviews of VL are now overwhelmingly positive as well.

The real credit goes to the developers.  I just try and report what I see.

[Lyn]

I think honest reviews that point out both the flaws and the benefits are always the way to go, if a community can't take constructive criticism then that distro is likely to be doomed.    Of course there will always be points where there reviewer will disagree with a philosophical decision of the distro makers but that is quite separate.  My pet loathing are the reviews where the main concentration is on the look of the distro while ignoring the nuts and bolts of of how the distribution works.

So in short keep up the good work, point out the flaws and praise the plus points.

Thank you, this honesty makes your reviews of all distros the one that I seek out if I want to know what the value of them really is.