It’s once again time for my biweekly Photojojo post. For those of you who haven’t been following my blog for a long time, Photojojo is a digital time capsule service. Every two weeks they send me an email that has my most interesting photos posted to flickr from one year ago.
The biggest takeaway which is SOOOOO crazy is how much Scarlett has changed in just 1 year. Only 365 days. And she looks COMPLETELY different!
I thought all my Photojojo entries for 2013 (consisting of 2012 photos) would be Scarlett, but then there was this screenshot I posted.
It was the 3 desktop backgrounds on the virtual desktops on my Main Activity. I was posting it in hopes of having it featured on a website.
Around a year ago I decided to KDE in Spanish to learn some technical terms. Back then I was using GDM, but now I’m using KDM. I didn’t see a way to set the language! How would I change the language to Spanish? I took a look online and found instructions. I’m going to reproduce them here for others. The great thing about the way that KDE handles things vs the way that Gnome does is that you can set a fallback language. When might this be useful? Let’s take a Vietnamese computer user. Vietnamese people (at least of a certain age) tend to be fluent in Vietnamese and French with some familiarity with English. So a Vietnamese person could set his computer to Vietnamese with a fallback to French for any programs that didn’t have translations into Vietnamese. As usual, I LOVE the level of customization in the KDE desktop.
And once you do that, each program launched from that point on will have those language options.
I did it! I have created my first ever useful GUI program. After all these years of thinking that all the useful GUIs had already been invented – I found an itch to scratch. And here’s how the final, working version looks:
And so there it is! I was actually surprised that this last little bit of the program wasn’t as hard to finish up as I thought it would be. So, as currently implemented, when the plasmoid loads up it makes the data engine fill itself. Once that’s complete you can click on the buttons along the top to load up all the photos from that group that are ready to be promoted to the next group. Clicking on a particular photo loads up the small thumbnail. View on flickr allows you to jump to the website in case you want or need to do that. Originally I implemented it so that if the next part was hard, I’d at least have the functionality of the command line version of this software. Then came the new part that happens to be easier in a GUI than on the commandline. You can add the photo to the next group and remove it from the current one. The reason I did it that way was so that if the photo couldn’t be added to the next group, it wouldn’t be removed from the current one. I tested it first with a fake group and then the real thing and it works fine. Before I post the code for my data engine and plasmoid, I have some cleaning up to do. I also need to add in a bit more functionality for robustness. After that, if I’m going to make it so that anyone can use it (and post it on kde-look.org), I will need to do some extra work. So that’s version 0.1. For version 0.2 here’s what I’d have on my TODO list:
make the add and remove buttons change color to green or red to indicate success or failure, respectively
add “Time until next flickr day” because these groups only let you add 5 photos per day, but flickr’s day resets some time in the evening (they’re probably on GMT)
Add a button force the data engine to refresh its data
Combine the add/remove to group code. I put it into two different helper programs but 99.9% of the code is the same. It’d be easier maintenance if they were in one program
Clean up the code by moving some of it to separate qml files that get imported in so it’s easier to read what’s going on
I’m quite excited and I plan to post the code, along with some explanations soon.
OK, so it’s been a year since I last blogged about working on this program. I tried working on the data engine in August and then got stuck right around the time I started my first semester of grad school so I had to drop it, even though it got stuck in my head and I was thinking about it for weeks until the light bulb went off. I wrote my idea and just had the idea to work on it recently. I have to say that, overall, QML is pretty awesome for making a quick GUI. I’ve always struggled with GUI code, but with QML I was able to put together a quick GUI in about 20 minutes. Now, don’t judge QML too harshly because my plasmoid looks ugly. It looks ugly because I just put together the minimum GUI to implement grabbing data from my Data Engine. Once I get everything working right, I’ll fix it up. After all, GUIs are really so easy in QML that it can be the icing on the cake in the end.
Man, when I got my plasmoid to this point I was soooo happy! Unfortunately, I wasted countless hours trying to figure out why my plasmoid couldn’t talk to my data engine and it turned out it was because I was using QT Creator (an AWESOME IDE for QML development) and for some reason using that instead of the plasmoidviewer commandline argument to launch my plasmoid means that it can’t talk to engines. It was so frustrating when I found out that was the reason for the issues. I had literally spent hours trying to figure that out. I’d say, use QT Creator to make your GUI and then maybe create some fake data you can read out of a file to make sure the GUI’s working. Then move to plasmoidviwer for your final development.
One other annoyance was that in the XmlListView I had to change had to change source: to xml: – another hour or so wasted from not quite understanding the documentation well enough. Basically it can only read xml from a data engine if the type is xml. If you use source, it only works if it’s a web page or a file on your computer. That’s kinda non-intuitive and wasted at least another hour of pointless debugging.
But I’m really happy because now I’m at the point where I’ve essentially ironed out enough of how QML and data engines work in order to finalize the look/feel of the plasmoid. Then, once that’s done I’ll need to get help from the IRC or the mailing list to iron out the final bugs. And after that I’ll post my code because there’s a real dearth of example code out there because QML and data engines are so new. And it’s not always clear from the code documentation exactly how things could fit together. Speaking of help, sreich on the plasma IRC room and Sebastian Kugler on the mailing list. They were very patient with me as I worked otu what was going wrong with my plasmoid. sreich even used his homework procrastination time to help me out.
Hopefully I’ll be able to have another post really soon where things are working at 80-90% of where I want them. It all depends on how much time I can find to work on the code.
So I took a look at Blogilo a few days ago. So, taking a look at my Blogilo post I have to say that it is pretty much ALMOST there for the perfect offline blogging tool for someone using WordPress. Sure, it doesn’t allow your to create categories, but a blog like mine that’s been around for ~7 years probably has all the categories it needs. The extra fidelity can come from tags which Blogilo hands just fine.
I mentioned that it doesn’t list the categories hierarchically, but that’s barely even an issue. The biggest issues for me are the lack of captions for images and the fact that uploaded images don’t automatically link to the images. In all of my images, I have them set to link to themselves so if you want to see it full-sized (pretty important when showing screenshots), you can. Blogilo doesn’t have that ability and that’s what pretty much kills it for me because that’s so important to me as a blogger. But I’ll talk to them on irc and/or submit a bug report for that feature.
I think if they just solve these little issues, Blogilo will be the BEST offline blogging software and might even give the WordPress native interface a run for its money.
For the most part, I haven’t seen the point of using blog editors like Blogilo. While there might have been a purpose to them back in the dialup days, it seems pointless in the days of always on broadband connections. Also, back before blog software like WordPress had their great visual editors, I could see the need. However, I guess I could see some use for it on my laptop. I often compose blog posts on there in a text editor when I’m traveling. It would be nice to have it all formatted correctly and ready to go when I got an internet connection rather than have to load up WordPress then spend time formatting it when I could have done that on my laptop while I didn’t have a connection.
Blogilo is developed by the same guy who does the KDE microblogging software Choqok, so I figured I’d check that out. So, here’s what Blogilo looks when you first open it up:
The neat thing is that you can put your website and username and password into the “add blog” dialog:
And it will automatically do all the configuration for you. And the last tab tells you what’s supported. Here’s what it looks like for my WordPress blog:
So far, here are the only negatives I see to blogilo:
It doesn’t have an easy way to add captions to images like you do in WordPress
The post categories aren’t listed hierarchically. While you can easily find them alphabetically, this can make it a little harder to see how they fit together
It does allow you to set the timestamp as well as whether you’ll allow comments and trackbacks. I think the neatest feature is the ability to see how the post will look on your blog:
Most offline editors don’t have something like that. Well, time to upload to see how well that works. I’ll be back tomorrow with my thoughts on the process so this can remain as uploaded by Blogilo.
I’ve been reading a lot about kde-telepathy and it seems to be the next evolution beyond Kopete. So I decided to check it out in my Fedora Rawhide VM since that’s going to have the latest packages and telepathy is still in deep beta. When I installed all the packages that seemed to be important, I got the following when I launched it from the alt-F2 menu.
So I decided to see what kinds of accounts were supported:
There weren’t any choices in there! It looks like it consists of plugins and I don’t have those installed. So I went back and installed the plugins. On the one hand, that’s good because your system isn’t too cluttered. You only need to install support for the networks you need. On the other hand, it means you have tons of packages to install. Hopefully whichever distros package it have a meta package that installs everything you need to be up and chatting. After doing that, I saw that they did indeed support tons of networks including Facebook.
I selected Google Talk and it was pretty awesome that it filled in all the important data for me. Although it’s not impossible to find, it’s annoying to have to hunt out the data for a particular network so it’s great to have the defaults filled in.
I loaded in my AOL IM selft and I was able to chat just fine. However, it was hard for me to figure out just what I had to do to get this chat window to come up. Apparently I needed to launch kde-telepathy-contact-list.
The neat thing is that the people who were renamed were done so on my other computer – so that data transfers back and forth. It doesn’t appear to have preserved my groupings or maybe that hasn’t been implemented yet.
So, it’s pretty much good enough right now for simple chatting, which is what most of us do most of the time. I didn’t test how well it integrates with notifications, but I’m sure that’ll be handled pretty well since it’s KDE tech. My only concern is all the fragmentation with the project. It’s great in that it allows each part of the program to be developed independently and swapped out without affecting the entire program. But it is confusing to know what packages to install or to launch. They’re going to need to do a good job of showing all the benefits and how it integrates with plasmoids and other programs in KDE in order to get people to see all the cool features they’re always blogging about. I’ll give it a deeper test once I upgrade my main Fedora computer at the end of this semester. Right now it seems ready for everyday usage, but only actual everyday usage can truly prove that. However, the team has made great progress and I’m sure by KDE 4.9 if not sooner it will be ready for the masses.
Back when I first started using Linux I was using a very underpowered computer that I got donated as part of my research at school. So OpenOffice.org was a real pain in the butt to use. It took forever to load! KOffice, on the other hand, loaded up quickly. At that time, with KOffice 1, they had the presentation program, the spreadsheet, and the word processing program.
At the time there were two neat things about KWord that I really liked and they both stemmed from something I was doing at the time. I was the head of the Phi Sigma Pi (scholastic frat) alumni team. We had a newsletter to get out. Turned out that KWord, as it worked at the time, allowed for very easy desktop publishing. I don’t think it was intended to replace Scribus, but when compared to Microsoft Word or OpenOffice.org Writer, it was MUCH easier to set the layout. That made it really easy to create. With MS Word (at least 2000 and 2003) it was SUCH a PITA to try and get images where you wanted to be and to have text properly wrap around it. We had all kinds of issues in this class I took my sophomore year in which we had a tutorial class session on how to do this and it was still horrible to work with. KWord also had the ability to output to PDF which was great as I was trying to help the frat save money by seeing if we could get the alumni to accept emails instead of letters. They ended up rebuffing that offer, but at least it was something I was able to do. (I hadn’t discovered cutePDF and other ways to create free PDFs on Windows)
Eventually I had to abandon KOffice in favor of OpenOffice.org because KOffice had horible compatibility issues with Microsoft Office. It’s a harsh fact that most people out there (in the USA, anyway) use MS Office and when they went to their XML file format, KOffice was no longer an option. In the meantime, The OASIS group came up with ODF. OpenOffice.org was quick to jump on that. I thought that all of the FLOSS would get there quickly too – it was a no-brainer. But, strangely it took Abiword and KOffice a while to get there. I think it may have to do with the fact that they were waiting for 1.1 or maybe, in the case of KOffice, they were busy getting ready for KOffice 2.0.
Now I’m no longer in school. (Haven’t been for quite some time now) KOffice 2.x supports ODF. OpenOffice.org supports ODF. I only share documents with my wife and if I have to send them to my parents I can always either save as doc or PDF. (Also, there are ODF interpreters for MS Office) So I can use whatever program I want for my office suite. I want to take a look at all that KOffice has on offer. However, since I’m no longer in school and don’t use KDE at work (nor can I take home work – corporate espionage fears and all that), I don’t care about KPresenter. I also almost never write anything that needs something as heady as KWord. So my main office program is the spreadsheet. I have spreadsheets for my workout weights, my time off (so I know in advance how much time off I’ll have at a given point in the year), and some other stuff. I’ll also take a look at Karbon14 vs Inkscape although I’m pretty happy with Inkscape for my work on my webcomic.
I figured I’d take a look at the wikipedia pages and the offical KOffice pages before trying them out to see if I can glean any important info that will help with my eval. The KWord article on Wikipedia was short, but confirmed that they are still using Frames. Taking a look at the KWord official site is pretty exciting. First of all, it looks at thought KOffice (or at least KWord) has been optimized for the fact that “everyone” has widescreen monitors now. It looks like they’ve used this extra space to the right to create a variant on MS Office’s ribbon interface. What I like about it is that it looks a lot less confusing than the way MS did it. The top bar isn’t constantly changing and important things like the “save” icon aren’t hidden. (Of course, I also take a moment to note that, like an expectation of high bandwidth – widescreen monitors may not be readily available everywhere)
KSpread’s Wikipedia page has some conflicting info on XLS support, but that’s not as important to me at this stage in my life. The official page shows that also looks pretty slick. The real test will be if it can import my ODS files from OpenOffice.org without any problems.
The KPresenter page on wikipedia is pretty much useless. The official page shows that it’s basically what you expect. Nicely, it does have a plugin architecture so if you’re doing presentations off your own machine you could probably find some nice plugins for special effects or something.
With Krita KOffice goes above and beyond other office suites and offers a bitmap (or raster) editor. The GIMP is also a bitmap editor. So I asked on Identi.ca and got the same information that’s on the wiki page and the official page. Basically, it’s intended to be for drawing and digital painting. So while The GIMP, like Photoshop, has to be a jack-of-all trades and, therefore, quite complicated (having to handle photography, image manipulation, and drawing), Krita can focus on the drawing part. I could see Krita potentially gaining more and more users among the web comic industry. (Windows compatibility is being worked on for the 2.3 release) After all, why use The GIMP or pay tons of money for Photoshop (which has PHOTO in the title) when you’d have this program perfectly tailored for artists? I’m looking forward to playing with this and recommending it in my web comic circles. (I use Blender for most of my art so I probably wouldn’t use it too much)
In fact, using Krita for drawing your webcomic would be great because KOffice also includes Karbon14, a vector drawing program. After all, vector programs are great for speech bubbles and text. So you could load your image from Krita into there and given that it’s all one office suite, probably have it automatically updating. The official page is here.
The last three parts of of KOffice are programs I’ll nearly never user. Kexi is an open source version of MS Access and much more. It can, for example, connect to a MySQL database. I have never used one of those because I’ve never had a task that was better suited to a database than a spreadsheet. (This may change as I setup a database for something my wife does – I think it might work better for her) KPlato is a project management tool. I usually don’t use those in my day-to-day life. I just don’t have projects that are large enough to be worth the effort. I used to use one when I was working on animation projects. Finally, there’s Kvivio a diagram program. But that one’s not even ready for KOffice 2 so I’m not going to look at it.
So here’s what it looks like when you start KWord. (Pretty much every part of KOffice looks like this when you start it)
I’m really conflicted about this. On the one hand, it’s kind of dumb that MS Office and OpenOffice.org open up a blank new document upon startup – what if you want to use a template or open a file. On the other hand, this stands in the way of getting started right away. Part of me thinks that the traditional way is better because you can just start writing right away. So I click on blank document. This gives me more choices.
I expected to be in Kword by now. I’m one of the first people to yell when we cater to “stupid users” but this is quite counterintuitive to what I thought would be here. To me, blank document means no templates. So I click on blank page. So here’s where it leaves me (and I’ve typed some nonsense)
So what do we have here? A great layout for a widescreen monitor. (I just happen to have it open on my left monitor, which is 4:3) I will tell you right now that this interface will probably scare the crap out of certain types of people – people afraid of change. There are a LOT of those people out there. That’s OK. Everyone doesn’t have to look like MS. It just means a shop going from Windows to KOffice will need to spend a bit of time and money on training. A neat thing for people participating in the Nanowrimo is that the word count is right there. Unlike other writing programs, you don’t have to keep going to a menu to check and see if you’re hitting your word count. So I played around with the other stuff to see how it works.
I learned that KWord also bucks another trend. To stop bullets you don’t hit enter twice and you don’t hit the bullets button again. You just backspace over the bullet. More intuitive to someone who’s never used a word processor. Less intuitive to those of us who have used them nearly our entire lives. Another thing I noticed is that it doesn’t appear to be doing live spell checking. Perhaps that’s an option that needs to be enabled.
Yup, it was off. Hmm….turning it on didn’t fix it automatically. Maybe I need to go back in? Hmm, that doesn’t appear to fix it either. Well, that’s more or less a deal-breaker. But let me see if perhaps I just don’t have the dictionaries installed. Well, I do have enchant installed. So now what? I asked identi.ca. We’ll see if they get an answer back to me before I publish this.
OK, onto KSpread. As I said before, this one is the most important one to me. First I open up a spreadsheet where I keep track of my vacation hours. I created this in OpenOffice.org’s Calc. I was not impressed with how it looked.
While all the into was there and the formula appeared to be working correctly (I didn’t check), it looked horrible. Looking at that for too long would really hurt my eyes. Why does it look so bad? Why are the rows so small? Let me check my workout weights spreadsheet:
Again with the lack of vertical space! It also doesn’t appear to preserve horizontal space either. Well, let’s see what happens if I start from scratch.
So that works. Not a good sign for interoperability, but at least if you make it in KSpread to begin with and only work in there you should be good.
KPresenter appears to need a lot more work. I just opened up a blank page, picked the second layout and did the most basic thing that’s in every slide show – bullet. They look horrible.
I’m sorry, but presentation software needs to handle bullets correctly. That’s something that’s in EVERY presentation. Let’s see how it does at presentations. At least that works correctly. Get the bullets fixed. Then we’ll worry about fading and special effects.
Krita works well. It’s somewhere between Tuxpaint and The GIMP. Looks like it would take some time to master and get really good with it.
Very promising. I hope it continues to get support.
I created a default page in Karbon.
After my experience this morning where it appeared that there was no way to put text into Karbon14, I was about to write it off as needing a lot more work. Turns out that it was hiding under the circle/triangle/square button on the bottom right. I understand this leads to a consistent experience across all of KOffice. I think there’s merit in that. But a vector graphics program is all about graphics. And so all the shapes shouldn’t be hidden like that! Neither should text as that can potentially be important. I also tried opening up an Inkscape SVG file. It appeared to open correctly – a very good thing.
Starting it up seemed more involved than I wanted to get into.
It appeared to have everything you’d need for a project management program. It would be way too hard to demo without a real project or a huge time commitment.
So what’s my verdict? It’s definitely a complicated one. On the one hand, it’s wonderful that KOffice includes all these programs. Their idea to use “shapes” and other such concepts is great. You’re really getting a nice, integrated office suite. Overall, the programs work as they should. KPresenter needs a lot more work. The others seem to be in good shape. (Other than the spelling issue) The biggest disappointment was with the interoperability. I understand there probably isn’t coding in the spec for cell size, but for it to look so atrocious. It just sucks. So try it out. It will most likely meet your needs after you get used to the interface. Hopefully it evolves so that programs like Karbon14 don’t have all the important stuff for that program hidden away just to maintain visual consistency.
Sure, it’s a tired and cliche phrase, but hurray for the wisdom of the crowd. I’ve received comments on identi.ca, twitter, and in the comments here with answers to nearly all my problems with KDE. Let’s see if I can get them all to work. First off, I was told that my problem with Konversation not getting my password in time to keep me from being signed into the fedora-unregistered could be solved by setting the password as a server password. Alright! That worked! woohoo! Before I’d had it set to just run the /msg identify command.
Second, getting Facebook to work in Kopete. Kofler, who also contributed the tip above, linked me to some pages. I decided to try this one first. Essentially you add a jabber account with the username of email@example.com and uncheck everything in the connection tab. yourfacebooklogin = your profile’s URL (minus the facebook.com part).
For Akgregator, I wanted to be able to see each of Dan’s posts inside Akgregator without launching a tab. Going to the advanced setting of his feed worked in getting it to load. However, it did not load videos. To do that I had to load it in a separate tab. So it somewhat works. If it’s text-only then you’re good. Otherwise….perhaps there’s some other setting to tick?
This next one’s a bit tricky. I had to hunt all over bugzilla to figure it out. In the summary page of Kontact I would see my TODO items duplicated. It would also print with duplicate TODO items. In the actual TODO page it was not duplicated. WTF? My clue was that when I added stuff to the TODO list it had two resources listed. So there were bugs from KDE 3.x where you had to delete duplicate entires in $HOME/.kde/share/config/kresources/notes/stdrc. But there was not a duplicate entry for me. So I finally found this bug that explained it. Comment #9 had the instructions I needed.
And then it worked! Yay! I’ll be sure to report back here if it doesn’t survive a reboot or something like that.
Another thing I mentioned last time was that it seemed at though having my dynamic playlist only look ten songs ahead seemed to mean I could come up with duplicate songs. In the week and one half that I’ve been using Amarok I’ve had two songs come up twice. That’s not too bad considering I’ve heard somewhere between 75 and 125 songs in that time. (Based on my daily avg on last.fm and my knowledge of how much I’ve been using it in the past 10 or so days) I ended up coming up with the following solution, I used a proportional bias with playcount equal to zero and set it to 80%. Why not 100%? I wanted to see how things were affected by me going through more and more of my playlist and leaving Amarok with less and less new songs.
I wanted to add another stipulation so that it would also tend to favor my most recently added songs. That way I could have more of a chance of hearing stuff I’ve just added to the library. Unfortunately, that did not appear to be a condition I could match against.
Well, that takes care of all the problems I was having so far. Thanks to everyone who responded in the comments and on identi.ca. I *did* find a bug with the TODO list where it can’t print more than one page worth of TODO items. That is very annoying. However, someone’s already filed a bug on it. I just added a “me too”. It didn’t seem to have much work on it. Perhaps no one else prints their TODO lists but the two of us? (Or at least no one with that many tasks) Next up I’ll be taking a look at how KOffice stacks up. With the big American Thanksgiving holiday coming up I probably won’t get to it until Monday, but I’ve already done a lot of the background work ahead of time so perhaps it will be up before then.
There was a time when I thought Amarok was the best music player on Linux. I even used to run it in Gnome as you can see from this 2005 screenshot. In that first link you can read me gushing over Amarok 1.4. I loved all the integrated technologies, especially the metadata juggling Amarok did. The first few Amarok 2.x releases with the KDE 4 libraries were complete crap. They were ugly and were missing nearly all of Amarok’s features. (Mirroring the complaints people were having about KDE 4 at the time) When I took a look at Amarok and KDE 4.4 in October I said I would take another look at Amarok.
Well, all this time KDE had been calling to me. There was something elegant and beautiful in KDE 4.4 that kept tugging at me. Maybe it was time to abandon Gnome again? I probably wouldn’t stop using all gtk apps – I love me some gPodder. But perhaps I would switch my major apps and desktop environment over to the KDE side. It would be the first time in a very long time that I did that. (Although with Gnome 3 coming, perhaps it’s the perfect timing for KDE’s new-found stability?) So after I did an upgrade to Fedora 14 that left Gnome’s panels a bit buggy, (I later realized it was *probably* my fault because I forgot to run yum distribution-synchronization – although as I write this I haven’t tested Gnome since running that command) I decided it was time to log into KDE and check out Amarok.
Amarok detected most, but not all of my new albums on the new Album widget. I didn’t check if the missing albums (new MC Frontalot songs) were in the library or not. But what it made me think was that I need to check and see if I can configure the widget to show more than just five albums. I mean look at all that empty space.
When I compare this to how it looked in Amarok 1.4 – you can see that the three panel layout makes more sense. Before you had to switch back and forth between the playlist and the context stuff that’s now handled with the widgets. The thing that’s LESS elegant in Amarok 2.3 is that these widgets are all separated whereas they’re stacked in Amarok 1.4. So you have to click around rather than getting all the album and track info at a glance. It allows for a more customized experience, but I hate having to click around.
When I got home I decided to play around with the widgets before I complained about it. After all, it was only fair! So I started playing around with the widgets and found that most of them could be configured if you right-clicked near the title of each widget. So for the albums applet I was able to up the number of albums via this setting:
And viola! It had more albums (including at least one of the missing MC Frontalot albums) See:
Also you may notice that it fixed the problem I was talking about with wasted space. Check this out.
Apparently, if you put the modules in the proper order – they’ll sort themselves out like this. There’s just one problem – I can’t figure out how exactly it works. Like can it only be three things? What determines how many things you can fit on one screen? It wasn’t quite intuitive to figure that out. Still, it is indeed neat to have the photos there. It’s just that if I put the album info above the photos, then the photos don’t show! Perhaps someone can explain in the comments?
Now, as you can see, the automatic ratings are working correctly here. The more I listen to a song, the higher it goes. If I skip songs, it lowers. I like that. It allows the computer to do a better job of being an objective observer than I can be. There are lots of songs that I have marked as 4 stars in Rhythmbox.
As a quick recap: I have at least one week if not two weeks of continuous music on my computer. I want to be able to just randomly play from this music without having to constantly skip songs I don’t like. So I have an auto-playlist (AKA smart playlists in some music programs) that grabs all songs rated greater than four stars. So I have to tag all my music. Untagged music is not heard by me. I recently either blogged or microblogged about discovering some albums I had COMPLETELY forgotten about because they had languished unrated. Of course, rating songs takes my attention away from other things. So I have to only rate songs when I’m doing a task that I don’t mind being interrupted from every 3-5 minutes to rate a song. Rhythmbox has one other flaw (depending on your point of view) – when you play songs on random, it gives greater weight to higher rated songs. Actually this reminds me of another flaw that I will mention in conjunction with this one. Just give me a second. So it’s not really random and I’m actually more likely to hear songs I’ve tagged as five stars (I REALLY like them). But the point of random is that I don’t want to always hear the same songs and get sick of them. This combines with the other flaw I just remembered – Rhythmbox does not save playlist state. In other words, let’s say I have a playlist of the newest songs I’ve added to my library. I start listening without random turned on. I get partway through the list and I go to bed. I will turn off my computer because I’m not going to use it again until I get home from work the next day somewhere around 16 hours or more later. I don’t want to waste money on electricity, harm the environment, or wear our my computer’s parts. The next day I boot into Fedora and turn on Rhythmbox and go to that list. If I hit play it will start from the beginning of that list again! Now, I can look on the right and figure out the last song I played yesterday and go from there, but that’s annoying. So if you combine both of these “flaws” – I tend to hear the same songs very often even though I use random.
Wow, that wasn’t a quick recap at all. Hope you’re still with me. Anyway, let’s get back to why it’s awesome that Amarok automatically rates my music. So, as I go through my collection, I know that if I rate something as 3 stars I’m effectively never going to hear it. Right now there’s an insane amount of music in my 4 star and up list. I don’t need more music in there. So the initiative is for me to rate a song at 4 stars if I think I’ll want to hear it again. (I *do* use 1 and 2 stars for songs I NEVER want to hear and songs I hate, respectively, in case I ever make a list of 3 stars and up) So there are songs that I may THINK I like just because I want to make sure I hear it. Or there may be songs I THINK I like because of emotional reasons – favorite song in HS, first song I slow danced to, etc But a majority of the time when it comes up in Rhtymbox I’ll skip it. An automatic rating system based on how much I skip a song will provide a more accurate rating of what I like. Additionally, as my music tastes change (as they have with my age), the ratings will evolve to reflect that.
Now, there was one thing that was making me hesitant with going back to Amarok. With Rhythmbox, Banshee, Songbird, iTunes, etc your entire music collection is already a playlist. You just start playing any track in your library and it keeps playing from there. But Amarok is based on playlists! I know I’ve talked somewhere on this blog (or maybe the old one) about how I hate making playlists. I get paralyzed about which songs to hear. Just like when my wife asks what I want for dinner and suddenly I can only think of about three of the dozens of recipes we know how to make, when I try to make a playlist I can’t remember from the hundreds of artists I have on my computer. So what to do?
Amarok’s dynamic playlists to the rescue! It generates a random playlist based on some criteria or on pure randomness. I went with the latter. I was happy with the results. It was truly random. I heard a Shakira song I hadn’t hear in ages. And, here’s the best part compared to Rhythmbox! I exited Amarok to reboot after a kernel update.
When I came back, it had found my spot and I was able to keep going. For once I will truly be able to go through my entire collection (more or less) and not hear the same song again until I’m done! Things are really looking up for Amarok!
Some other features I noticed that I liked include the ability to love tracks on last.fm from within Amarok. As far as I know, Rhythmbox ONLY allows this if you’re listening to a last.fm stream. With Amarok you can do it while listening to your library. It’s not a killer feature on its own, but I know that I have “loved” far fewer songs than otherwise because, right now, I have to go to the last.fm website to do that. So that’s something I really like.
Another feature that has finally made it back from the 1.4 days and is working properly again is the ability to set playback to stop after a certain song has finished. This doesn’t happen THAT often, but there are times when I want to finish listening to a song as I get dressed or do something else, and then have the playback stop without me having to go back to the computer to stop it.
One quick thing I wanted to address specifically for Dan. I mentioned him before as another huge fan of Amarok from the 1.4 days. He said that he can’t use it because it messes up his Japanese songs and the metadata is just squares. Dan, check out Sambomaster on Amarok:
Not only that, but the upcoming events even works!
Is there still room for improvement? Sure! Perhaps fixing the widgets up a little more. Improving the ability to scan the library more quickly. Really the danger is that it’s finally so good that the devs will get bored and either bloat it up with useless features or let it decay.
I have decided to stick to KDE for a week or two to properly assess whether I will migrate back to KDE from Gnome. So in that time I’ll be using Amarok for music playback. That should give me a lot more time to notice any niggles that annoy me as well as find any stability problems.
For a quick chuckle…I don’t know what Amarok is using for the backend for grabbing the band photos, but there was a bit of a FAIL with the old 1950s/1960s band The Playmates
For some reason, I didn’t get Linux Format Magazine issue #110 when I was supposed to. I ordered another copy and it arrived recently, so it’s time for another slate of Linux reviews. Unfortunately, something appears to be wrong with the way they mastered the magazine DVD, because I was unable to boot into any of the Slax options. So I went online and got the latest ISO off of http://www.slax.org.
Slackware was the original Ubuntu. It was the first very popular Linux distro and it is the oldest distro still in production. Most users have on to Red Hat, Mandriva, Gentoo and, finally, Ubuntu, but Slackware still exists. There are a lot of people out there who still use it. I don’t know if this is because they cut their teeth on Slackware back in the day or because they like the pain of doing everything by hand. You have to admit there’s something nice about Slackware’s programs being the most untainted by the distro. You know that Slackware’s KDE is the closest to bare metal KDE you can get.
Strangely, considering how old Slackware is and how it’s considered the Linux distro for Linux experts, there are a lot of distros based off of Slackware. Slax is a LiveCD based off of Slackware and it’s been getting a lot of attention recently. Some of it has been good and some not so good.
I loaded it into RAM and it runs just as fast, if not faster than my installed Linux distros. Interestingly, compared to other Lightweight Linux distros I’ve reviewed it uses KDE instead of Fluxbox, JWM, or others. Here’s the default desktop upon first boot:
worked without any extra configuration
Slax basically consists only of KDE programs. Of course, this is fine because KDE programs are going to be well known (unlike mtpaint) and a lot of them are at or near the top of their class. Kolourpaint is for creating raster graphics. Kate and Kjots are available for text editing/programming. The KOffice programs are available for Office-level projects.
Kopete is included for instant messaging. Konqeror is the browswer (which is why I don’t have screenshots – it doesn’t work well with WordPress). Kmail for mail, Akregator for RSS feeds, and other tools for the net. Very comprehensive suite.
For multimedia applications we have Juk for audio and Kplayer for video. Gaming consists of KBounce (a very addictive Quix-like game), Patience and KBattleship.
So, last time around, I said that Antix was the new king of Lightweight Linux distros. Does Slax unseat it? On the one hand, Slax has KDE as the base system so it’s automatically going to be more familiar to newer Linux users. On the other hand, the choice of KDE probably means it won’t run on computers that are as old as the ones Feather, Puppy, and Antix run on. Still, there’s room for a distro like Slax for recently obsoleted machines. And, if you want to run it on a 3 year old machine like mine – it flies. Now, some of the other lightweight Linux distros have more programs or DVD ripping software included. However, this brings us to one of the neat aspects of Slax – modules.
While many distros and LiveCD distros are incorporating similar features, Slax is still unique in that you can remix the LiveCD to include whatever programs you want. You just go to the Slax website and download the modules you want. Then you put them into a certain folder and burn a new CD. Bam! Now you have a LiveCD distro with exactly the programs you like….as long as they exist as modules. They have a lot of work to do to get more programs working as modules, but, as far as I can tell, this is a new feature with version 6 so they can be forgiven for missing some programs.
Overall, I’m very impressed with how fast it runs if you have the necessary RAM. The program selection is decent – could be a little better. I would have liked Firefox because Konqeror doesn’t work quite right for me on all sites. I’d say that Slax is tied with Antix. I really like a lot about how Antix works and it truly is lightweight since it runs Fluxbox. Slax, however, has all that KDE has to offer and modules. If I were to walk around with two CDs (I don’t even walk around with one), I’d be sure to have Antix and Slax around for use on any computer where I wanted to run Linux and then leave without a trace of it. One more thing in Slax’s favor – like DSL, you can download files to easily create a bootable thumb drive/memory stick to use if your BIOS supports booting from USB. Definitely check out Slax, you’ll rethink your image of Slackware as a hard to use distro.
Compiz-Fusion, as you surely know is responsible for eye candy on GNU/Linux distros such as windows that turn into paper airplanes when the user minimizes it to turning the desktop into a spinning cube. There’s something about the wobbly windows that provides some a sense of inertia that just makes things feel a little more dynamic on the desktop. I can’t explain it, but some of the effects make the GUI slightly more useful. However, Compiz-Fusion isn’t perfect.
When I’m using Compiz-Fusion in Gnome, my system try notifications from Rhythmbox don’t work. Also, when I’m in KDE, Compiz-Fusion royally screws things up. (At least it does on my Fedora 8 desktop – your mileage may vary) It ignores my themes, messes up text on some of the windows and generally makes KDE crash even more than usual.
On top of these bugs, KDE 4 now has its own built-in composite manager capable of doing the spinning cubes and Expose-like desktop bling. So with KDE, Compiz-Fusion is now redundant instead of required for making the desktop look nice. At any rate, it was a bit kludgey with KDE to begin with. There are a few, though not many, distros that use KDE as the main (or, in some instances, only) desktop so those will presumably get rid of Compiz-Fusion once KDE 4.1 comes out this July.
On top of this, I’ve read that work is going into Metacity, Gnome’s default Window Manager, to add more compositing features. I’ve seen some screenshots of transparency working already. They’ll be working on spinning cubes, etc soon, I’m sure.
I think that both KDE and Gnome were going to get around to true compositing (including true transparency) eventually, but they were waiting for the X.org composite features to mature a bit more. Some programmers got impatient and thus created Compiz and Beryl which then merged into Compiz-Fusion. (And some smaller window managers like Fluxbox had already started work on this before the big boys) In fact, even XFCE now has some small compositing enabled. (Not too much since it’s supposed to be a light-weight desktop environment)
I know that Linux programmers don’t always work together in the best way, but I think that eventually Compiz-Fusion will be made irrelevant by the main Window Managers/Destktop Environments including compositing. Then, if the Linux programmers work together well, I see Compiz evolving into a set of plugins for Metacity and KWin. So Metacity and KWin will have the basic eye candy built into it such as transparency and Expose. If the user wants crazy stuff like Windows burning up when they’re minimized, they can then use these plugins. Only time will tell if I’m right….