Sunday, April 18, 2010

Returning to Linux, Round 2...Mint

So as I got some feedback from my last review, a couple of oddities...first, the developer himself told me that the distro had locked repositories...ummm...why would you ever put anything out and then support for only a few months? This is why I have an issue with the hundreds of distros out there, many of which are just recycled parts of main systems, removing Firefox due to hated of copyright...removing all shreds of anything proprietary...so that brought me back to thinking, well who had ever done this right? A couple of thoughts, first, PCLinuxOS, where Tex stripped and cleaned Mandriva like a Texas steer on a BBQ pit. Alas, I already gave them a go and will be waiting for everything to exit Beta...Then, MEPIS, another solid package where someone took bloat and made it work. Maybe, but never really 'loved' what was happening there.

Based on my efforts to get to a XFCE version that I would want more permanently, I looked back to what I enjoyed the most, researched again what was out there...seems that my first instinct of Linux Mint being something I should look back at was itching at my scalp like a long day on the golf course with no hat...

I had reviewed Mint previous, way back with Bianca...seen here with my archive on google docs. I had a great interview with Clem, it was read at least 5 times :) but I did enjoy the package back then. I saw recently that the community edition of XFCE had been released, so I downloaded and jumped through the basics of installing via the standard Ubuntu flair.

The Good
I always likes the ideas within Mint...make a good package better by adding little features and taking out what isn't necessary. I remembered from the day that everything just worked. Websites played flash, video on DVD would play with audio, documents opened without too much effort or downloading extras. So what else is here that most people would like to know about?















I love the update manager. This really helps people know what is going on, where relying on synaptic might just lead to confusion. By telling people level of changes, what the old was and what the new will be just increases what was done pretty well already.




















Software manager is even one step farther ahead than update manager. This now helps people know what is available, gives linux users a way to promote software they like, thrash that which is really useless and give people an idea of what is generally available. I love what Mint team has done here. By showing off the strength of linux, being able to add/delete/modify that which we have or need to have, users control their own systems in ways that are just not capable in either windows or apple. This actually reminds me of what Linspire had years back with CNR, yet was never able to cash in like they had hoped.

The not so bad
I guess money helps everyone and I know bandwidth demands cost. So with some levels of success comes underlying costs, a lesson I learned when my limited blogging success ran up a $2500 bill on godaddy. So while I never mind seeing donation links, they are how most get by (unless they are partnered with some large company, which then has profits in mind). I was taken a bit back by the splash window. I love that the discover links are there, however the concept of donating on launch...eh.
















I guess then when I went into some of the pages, I was even a bit more taken back from the idea of showing donations to a level of, well, a southern church showing off their needs on Sundays.
















I guess if you don't mind giving your name why should it bother me? Reality is that it doesn't bother me that much, however I just see more on their websites on this than I really saw on how to make things work better. Sure, the forums are pretty solid. I just would rather see this in the forums rather than so prevalent.
















I guess though this concept is working well. According to this Clem & Co. are getting close to $4,000 a month in donations. Again, I don't see this as a huge issue, I just would rather see this more internal to people who have signed up versus having a banner ad on the home page telling users some random donator and their level of how they compare to others. Again, I am reminded a bit too much of churches with slick preachers, slicker facilities and more on the church's needs versus how people can help people out in the community first. Preaching over...time to get on with the packages.


The tools for XFCE
XFCE is not the most efficient system. It doesn't have a traditional control panel like other packages. It has little packages within a settings control, so new users may not find pieces, especially the most common. Once again, Mint has a solution for this.
 































These simple tabs control majority of what a typical user might want to change, yet within the limited controls of XFCE would have a difficult time getting these taken care of.

Wrapping up
I looked back at my review of Bianca, now that Mint has moved on seven times over, the main question was have they really moved forward? The answer at least for the community version of XFCE is yes. Having the tweaks and tools they do improve the package and they don't try to redo so much that the base is corrupted. The community has grown by leaps and bounds and I am sure Clem is now focused more on this that his old day job. My hat is off to the team and this is a definite keeper on my main partition.

Cheers!

Klg.

Thursday, April 08, 2010

Ubuntu Switches to Google...again...

I see that Yahoo! is getting replaced by another party. First, my former colleagues at T-Mobile realized how much Yahoo! was losing them in content sales and not producing in search revenue, now I see Ubuntu is probably realizing the same thing...or...they are realizing that since the Yahoo! Bing deal has passed federal acceptance, they now realize that Microsoft, and not some non-evil empire will be giving results. Although, since the Bing folks are also my former colleagues, I can honestly state Bing and Microsoft are just related remotely. Sure, there are some old school has-beens in the Bing group, but once they hired in more talent from other search companies and bought a few others, the ideals in Bellevue (where the OSD group/Bing is located) is quite different than in Redmond. Sure, it is only 20 minutes away in traffic, but the leadership at Bing understands it isn't about brute force.

The money part, well that I can understand. Having gone head to head with Google on search with Mobile, I can honestly say that Google's pockets are way deeper than Microsoft...shocking isn't? I know what MSFT paid Dell and HP for their placement, I know what MSFT paid Verizon for that deal, however I am not going to publish the numbers for many reasons (I can spell NDA and LCA), I know that in both cases Google was offering more. What MSFT brought to desktop were most likely (I didn't do that deal myself, so guessing) concessions on other items. For mobile, we offered a flexible platform, that would continue to promote content, ringtones, games and internal feeds, where Google treats all data on its relevancy. Why does that matter?

In Mobile, it isn't yet about search ad revenue, there just are not enough clicks and users generally to get the CPS model to a high value. Users just don't have enough space on the screens to be bothered by enough ads to get people interested in clicking them. So content, albeit a slowly decreasing and since Apple/Android started their own markets, just make more money than what search can offer. Verizon makes more on content in one year that all the revenue that could be offered over many years from search. Rumors were that Yahoo! cost T-Mobile 25% of sales on content. This would put it in the magnitude of $100-200 million a year. I would estimate even on a good solid year that Yahoo! might have done $5-10 million in ads in that time frame, possibly up to $20 million if they did something amazing...they didn't. That is reserved for the iPhone.

So what is left? Since Yahoo! killed off a majority of their search group with the Microsoft partnership, I would imagine more groups will go to Google. Yahoo! owned mobile space, with at least 10x all the other partnerships combined. Where are those deals now? My bet would be towards Google as I know what Microsoft can handle and willing to step up for...Sorry Yahoo!Bing, seems that open source will be closed to you, not that it really matters, your 24% just doesn't grow there.

Sunday, April 04, 2010

Returning to Linux, where to start? Part 1

I used to review linux, some of you may remember either knolinux or new2linux websites. Every once and a while a former reader would email and ask, "what happened?" How do you explain certain things in life? It wasn't health, family, ideological changes, or anything even close to resembling something dramatic. I simply took a new job...at Microsoft. No, I am not a programmer, so it wasn't even remotely related to operating systems, office, or anything that makes them money. It was mobile search, so in fact the pieces I worked on actually cost them several hundreds of millions of dollars. I digress. Since I left them after my 18 months of pain, shame and learning how lame it really is to not be a programmer, I have been working ever since on getting back into my engineering discipline and my love for linux. No, I wasn't forced to stop using Linux, I simply made a decision to try to understand my company's products, as I had spent nearly a decade of trying to make sense of what came out of Redmond. I only take from my time there a learning of how not to manage products, how not to spend money going after lost causes simply to try and beat others, and that free soda just doesn't make up for the slow and steady pull of life from the lack of color in the offices. 

So my last few months have been dealing with a physical move and at the same time learning a new role, which is actually an old one. At the same time, my poor old laptop was struggling with the load of Windows 7, as it is stuck with 1 gig of RAM (odd shaped slots prevent me from upgrading). So every once and a while I would download a new Live CD of a various distro that I was used to in  order to determine what was I going to head back too. Would I be a lemming and jump onto the Ubuntu ship? Go with what I knew best with Mandriva or OpenSuse? Or should I venture to some of my favorite distro's I reviewed previously to see where they all lied?

The one thing I knew is that I still can't stand KDE and Gnome has really become more and more about others dictating form and function. My two favorites from years past had really gone in two different streams: XFCE was becoming main stream while Enlightenment was struggling. I took a quick gander on Distrowatch to see what new and uprisings were happening on the leaderboard and noticed two old KDE camps both had a XFCE distro out; Mint and PCLinuxOS. I had rather enjoyed both and actually interviewed Tex and Clem for my previous reviews, so why not head back to a reasonably comfortable place. 

Now the first thing you might notice, I am not going to be doing reviews like I had done in the past. Those were a lot of effort of VMWare to go through the loading and installation. I figure that enough time has past and most distro's now have excellent forums, so my efforts would really be a waste of time. I decided to focus instead on what I think is done right, what needs some tweaking and are the distros really worth effort of installation for others. 


The first one I installed was PCLinuxOS XFCE Phoenix Edition.














The style is quick to impress, I love the look of flames on a wallpaper. The rest, well, let's just say for a first run at an XFCE, my expectations were not over or underwhelmed. 

The installation process had a hickup, in that it couldn't understand my partitions. They are not all that confusing and no other version I played with has run into any problems. I think it might be the fact I have an extended partition in the middle, making the system choke when running a expect. This wasn't too big of a deal, minus when I tried to load the other partitions up as their own folders (easier to find files from the others) the system crashed and I had to start over. So when I went into the pretty PCLinuxOS Control Center and tried to add these via  "manage disk partitions", it failed to mount the new drives. 















So then I tried to mount as root and it told me that it cannot recognize Ext 4 partitions...huh? Have I missed something over the past 3 years? Maybe a new magic command that I just don't know about? If someone could fill me in, would greatly appreciate it.


















So this really bummed me out, as now I am missing some files from this distro from previous screen shots and will have to install another distro just to gather them back. Not going too well so far. 

Other disappointment was that openoffice was not installed. I understand that this is a memory hog, however many others have found a way to install these without going over the magic 700 MB iso. I did notice in the web browser that there is now a Appstore. Like many other distros, it seems PCLinux now has a way to gather applications outside of Synaptic...why? I don't see the need to go beyond a solid method of gathering applications. One thing I did learn in developing applications at MSFT, KISS (keep it simple stupid). Why have two points of failure that are doing the same thing?















So what do you get as options? If you don't install office, wouldn't you think that office would be an option? Nope...just Opera, google chrome, Dillo, Epiphany...a couple of HTML editors, a music manager...eh...I just don't get it. Installing the applications actually seem to take longer than download and installing command line. I get the ease, however simply instruct people how to get all applications. In fact, my favorite was when Google Chrome didn't load, and I got this error message....classic.





The rest of the package I would rate as average. Nothing really screams wow, nothing else is really lacking. With this you get a solid XFCE distro, albeit a bit fragile that is not as robust as others. If you are a fan of PCLinuxOS, the support is there, the community that is second to none and the packages are customized towards what those users are wanting to see. I love that the links are there in Firefox and I love the line "The distro-hopper-stopper" as I have always felt PCLinuxOS was one that would stick with the user a bit more than others.


 












So how would I fix pieces? It really doesn't need a lot. Understanding a bit more on the installation issues would be a good start and finding a way to put a script to allow users to install some sort of office script would also be a solid step to help people get what they need with a system. Video playback on websites was solid, media is well represented. Like the Phoenix in the wallpaper, this is a nice picture, just not a whole lot of extra color.  I will keep it around for a while and that usually says that I don't hate the distro, I have hope, and patience so time will tell how long it stays.

So there you have it folks, my first review after a few years off trying to figure out how not to lose millions :) Let me know what you think and please offer suggestions of how to improve, what would be a good next distro to review or anything else on your mind. 

Cheers! KLG