I have read a great deal over the past few days about a "Fork" in OpenOffice.org. First I would like to say I am appalled at Groklaw's reaction to Novell's announcement.
http://www.groklaw.net/article.php?story=20061204130954610
Everything they are doing is being released back to the OpenOffice project, and they have maintained a separate version of OpenOffice for some time that they make widely available. Other distros do the same. Not only that but they show Novell's press release in the article where they accuse them of "forking" the project and in the press release is the following quote:
Novell will release the code to integrate the Open XML format into its product as open source and submit it for inclusion in the OpenOffice.org project. As a result, end users will be able to more easily share files between Microsoft Office and OpenOffice.org, as documents will better maintain consistent formats, formulas and style templates across the two office productivity suites.
Now I have never taken a law class, but I have been reading open source articles, and articles that take court cases and bring them down to a common level of understanding for a long time. Last I checked none of the words in this quote had been dramatically changed in definition by Websters. How Groklaw could write the article they did, accusing Novell of the things they have accused them of while placing this text in the exact same document is beyond me.
Then I read a bunch of fired up forum posts one of which linked to Luis Villa's Blog
http://tieguy.org/blog/2006/12/05/quick-thoughts-on-novell-blogs-as-journalism-etc/
He had a very even headed response to the entire thing, and feels about Groklaw about the same way I do. However he made one other point that I heartily disagree with. That is he says he thinks supporing Microsoft OpenXML is a bad idea because it gives legitimacy to the standard. Now I'm sorry, but one of the only things that has made OpenOffice a viable alternative to Microsoft Office is that when some average joe consumer sends you a file they made in Microsoft Office we have to be able to open it. So when Microsoft has a closed format it's appropriate to crack it, but when they make an open format and submit it as a standard it's unacceptable to use? Aren't we getting just a little bit childish here. If Microsoft changed their .doc format in an attempt to keep us from being able to read the files, or just because they wanted to, we would accuse them of playing unfair. We would have cracked that baby faster than my little sister used to break into the cupboard to eat all the raw sugar. Yet now they open it to use, as many flaws and absurdities as it may posses, and we say, no I'm sorry we wouldn't want to give you legitimacy.
Microsoft is not the anti-Christ. They are a corporation that happens to have a great deal in common with IBM from back in the day. Now who is one of our greatest supporters? IBM. Microsoft has some shady dealings, and many of the people who have made them despicable over the years are still there, but some of those people are fading out. Some day Microsoft is going to wake up and they are going to have as close to a functional management as a large corporation can have, and they are going to enter into a key meeting where they decide to do the next version of Windows or Office in a manner which is in some major way directly beneficial to Open Source. They will still be competition, but as with the Novell and Microsoft deal sometimes you have to work with your competition while still competing against them to get things done.
It's time for us to at least try and look at Microsoft's actions for what they are and at least turn the anti-Christ, nothing they do can be good tint down a little bit. If we don't the other 90-95% of the population that doesn't bow at the foot of the Open Source religion is never going to take us seriously.
Forget giving legitimacy, or who's evil. Just ask the people what they want, and then show them that Open Source can give it to them.
Wednesday, December 06, 2006
Wednesday, November 08, 2006
Internet Multimedia and Mozilla Flash Agreement
There has been a lot of madness going on recently in the open source community. Oracle grabbing up Red Hat's distribution and re-branding it as unbreakable. Which in my opinion is kind of funny, anyone who's ever tried to administer any computer system at all knows that the word "unbreakable" is kind of like a Greek woman fawning all over their new born baby in front of a shrine to Zeus.
In the midst of all this excitement there was a small insignificant announcement that Adobe was going to be handing over the code for the ActionScript virtual machine to the Mozilla foundation. This will have the effect of improving the efficiency of Flash content in Firefox by building some of the interpretation into the browser. This will certainly help with cross platform compatibility. Considering the headache they've been going through trying to implement Flash in linux using only a minimal set of libraries to avoid dependency issues it comes as no surprise to me that they made this move.
Personally I think this move in the long run has the potential to be more exciting than either of the "Take down the Red Hat" announcements. If this code is released under a full open source license, which it is my understanding that it will be, then there is the possibility of someone really attacking the project of writing an open source program that does what Flash does. Now I don't mean writing a flash replacement. That is such an enormous monolithic project I don't see how anyone could catch up to it. However, as with Photoshop there is only a small percentage of the people who use Flash who make use of all of it's capabilities. If somehow over the next several years someone was able to create a project that implemented vector graphics interactivity and animation in the Flash style and only got to the point of the Flash 4 feature set then that would be really exciting. I don't think it would cut into Adobe's market all that much either. The people who are willing to pay Adobe's astronomical pricetag for Flash aren't going to be interested in a smaller open source alternative. For people interested in just doing web interactivity work in the .swf format with Actionscript controls though having an affordable alternative would be a major step forward. Macromedia made Flash available at reasonable educational prices. I knew students who invested 100 bucks and got the entire MX 2004 suite from my college. The academic price for flash itself is now a couple hundred bucks. I wish I had the programming know how to start a project like this, because it is time. There is a huge market of people who are ready to make the web far more interactive than most of it is now, but aren't that crusty top layer that Adobe sells to. It's time to make a product for the rest of us.
In the midst of all this excitement there was a small insignificant announcement that Adobe was going to be handing over the code for the ActionScript virtual machine to the Mozilla foundation. This will have the effect of improving the efficiency of Flash content in Firefox by building some of the interpretation into the browser. This will certainly help with cross platform compatibility. Considering the headache they've been going through trying to implement Flash in linux using only a minimal set of libraries to avoid dependency issues it comes as no surprise to me that they made this move.
Personally I think this move in the long run has the potential to be more exciting than either of the "Take down the Red Hat" announcements. If this code is released under a full open source license, which it is my understanding that it will be, then there is the possibility of someone really attacking the project of writing an open source program that does what Flash does. Now I don't mean writing a flash replacement. That is such an enormous monolithic project I don't see how anyone could catch up to it. However, as with Photoshop there is only a small percentage of the people who use Flash who make use of all of it's capabilities. If somehow over the next several years someone was able to create a project that implemented vector graphics interactivity and animation in the Flash style and only got to the point of the Flash 4 feature set then that would be really exciting. I don't think it would cut into Adobe's market all that much either. The people who are willing to pay Adobe's astronomical pricetag for Flash aren't going to be interested in a smaller open source alternative. For people interested in just doing web interactivity work in the .swf format with Actionscript controls though having an affordable alternative would be a major step forward. Macromedia made Flash available at reasonable educational prices. I knew students who invested 100 bucks and got the entire MX 2004 suite from my college. The academic price for flash itself is now a couple hundred bucks. I wish I had the programming know how to start a project like this, because it is time. There is a huge market of people who are ready to make the web far more interactive than most of it is now, but aren't that crusty top layer that Adobe sells to. It's time to make a product for the rest of us.
Saturday, October 28, 2006
Xubuntu's Edge First Impressions
So as soon as the Xubuntu page updated, which was sadly after both Ubuntu and Kubuntu, I ran in and grabbed the shiny new 6.10 iso. I have not played with any pre-releases or betas, or production spins for the simple reason that I don't currently have a computer that is working. I was lucky enough that my boyfriend is getting a new laptop and turning over his Dell desktop that he picked up for 10 bucks because of a financial aid program to me. I can have a samba and apache server again, and it makes me happy. We are still waiting for his laptop to arrive, so this is only going to be a review of the live CD experience. Now I should preface this by saying I never really used Dapper. I used 5.10 Kubuntu and at one point did the Synaptic "Update All" thing. The interface changed quite a bit, so I presume this was basically Dapper, but it had some personality problems. So if I mention some issues I had with previous versions of Kubuntu in my case that seem well obscure and weird, consider the source.
So as soon as the Xubuntu page updated, which was sadly after both Ubuntu and Kubuntu, I ran in and grabbed the shiny new 6.10 iso. I have not played with any pre-releases or betas, or production spins for the simple reason that I don't currently have a computer that is working. I was lucky enough that my boyfriend is getting a new laptop and turning over his Dell desktop that he picked up for 10 bucks because of a financial aid program to me. I can have a samba and apache server again, and it makes me happy. We are still waiting for his laptop to arrive, so this is only going to be a review of the live CD experience. Now I should preface this by saying I never really used Dapper. I used 5.10 Kubuntu and at one point did the Synaptic "Update All" thing. The interface changed quite a bit, so I presume this was basically Dapper, but it had some personality problems. So if I mention some issues I had with previous versions of Kubuntu in my case that seem well obscure and weird, consider the source.
Bootup: Bootup was delightful. It was exceptionally quick for a live CD. I am looking forward to seeing how quickly it boots when actually installed. When the desktop showed up there was an icon for my iomega USB hard drive, which has been on the Dell since my computer toasted itself a couple weeks ago. There was also an icon for the DVD burning drive that had a random ATI driver disk in it that I hadn't taken out before I booted. There was no icon for the CD burner that I was running the live CD out of, but that makes a certain amount of sense.
First Action: The first thing I did was double click on the iomega drive and try to load a pdf and a .mpg. The reason I did this is my Kubuntu file manager always freaked out a little over actually launching files from the media:/ location. I hated the fact that if I wanted to run anything, I had to manually go to the location in the Unix file structure. I really wanted the drive icons on my desktop to go there for me since the media:/ thing obviously didn't work. In Xubuntu this worked perfectly, and again even though I was on a live cd the launch time was tolerable. Not what I'd call peppy, but tolerable, which is an excellent sign when launching from CD.
Icon Options and Arrangement: The Applications menu was well laid out. The options were in good places, and there was a decent selection. I had a couple problems. One there was no burning program that I could find. Now this may be because it's a live CD, but I would hope that since live CD function is generally used to evaluate an OS before you actually install everything would be here. There are companies now that you literally cannot purchase a computer from without at least a CD writer. As I have been shopping for a laptop to replace my now defunct desktop most configurations start with a DVD player/CD burner combo drive. I know Xubuntu is meant for older computers, but given how most software is distributed for linux not having the burning interface installed is kinda silly.
The Little Things I Noticed: Firefox is included with its honest to god, not free as in freedom icon. I say good for the Ubuntu people. I personally think that if Firefox wants to maintain a certain amount of quality control around the code that gets distributed under their trademark then they should. It is that kind of attention to quality that keeps so much proprietary software ahead of the game (go ahead flame me I can take it), and it's what has allowed Firefox to take so much of IE's market share. I personally hope they keep it up.
The next little thing I noticed was the "Add Remove" icon in the "Other" sub-menu of the Applications menu. Now when you open this thing you realize it's synaptic with the menus and options taken away. It even uses the same icon. I opened synaptic next to it to see if they were similar, and they do not have even remotely similar interfaces. In fact there is one interface change in synaptic from my old Kubuntu 5.10 install that I am rather annoyed about. The lack of a search field in the main window interface (something Add/Remove has I must say). When using synaptic you're probably going to be searching, because if you know the package that's what apt-get is for. The advantage of synaptic is you can browse package lists. I think it's good that they included Add/Remove. Personally I think they should have taken some time to make a different icon so they could both be in the system sub-menu. I mean "other" why would you go to all the trouble of including and possibly developing (I don't know if the ubuntu people did this interface or not) a whole interface for newbies, and then put the link to it in some weird place no newbie would ever look? This isn't the end of the world, but it's a sloppy little tidbit on the end of an otherwise nice move towards non linux usability while maintaining the options for power users.
Other Usability Complaints: Ok while I'm on the subject of usability. I would like to comment on the media:/ thing again. So while writing this review I open Abi-word to check out how it inputs .odt files. I click the open icon. It gives me a very attractive maclike interface. I click on the desktop to get to my USB drive, and am greeted with . . . nothing. There is no link on my desktop to my iomega USB hard drive. This is totally unacceptable. Ubuntu took the time to generate a whole false hierarchy that only works when you are in their file managers. They didn't even both to alter the code to the packages they distributed so that it would work with them. Way back in the day when I used Xandros 2.0 their weird "My Computeresq" file structure was at least recognized by the programs that Xandros supported. If you installed someone else's software you were out of luck, but it worked with what came with the OS. Now here is my question. If you are going to spend all that time developing a whole new data file structure, you might as well build it into the actual file structure. I mean really. I would rather the mount points actually be on the desktop. Would that be so bad, I mean really would it? Or here's another possibilty could you possible setup a program that could dynamically create actual file structure virtual links? Would this create a lot of messy referencing if you are in the terminal. Why yes yes it would, and let's see this is for the benefit of newbies. If a newbie is in terminal then he deserves to have to start actually learning what things are. This NEEDS to be fixed. I like the link on the desktop, I like that fact that I can actually launch from it. It absolutely must show up when I use the open file command inside of the programs that come with the flippin distro. My other problem is that the live CD did not detect my sound card properly. It is an on board Dell Sound Card. This is not a new dell, this is a rather old dell. It is running 256 megs of SDRAM, the 133 Mhz variety and a 1.6 Ghz P4. So it's not like there were any surprises waiting.
Ok now that my rant is done, I will resolve. This is a good distro. As I never really played with dapper I can't say how much of an improvement it is, but I can say it is closer to a usable desktop than any other I have used. My major complaints detailed above are shared by the vast majority of linux desktops. While Xandros has done a wonderful job of tackling some of these usability issues when you boot into Xandros even now it feels like you are running something that is almost a decade old. Is Xandros ready for prime time commercial desktop, sadly no. I have yet to play with a linux desktop that is. Is it a good choice for linux geeks wanting a complete distribution based off simple low system requirements packages that provide support for all the modern goodness that we have come to expect from distributions, definitely. As someone who doesn't care about bouncing icons and 3D animated squares for my desktop (it looks cool, but I'm sorry it's the stupidest UI idea I've ever seen), Xubuntu makes me very happy. Hopefully it will continue to improve over time, and in a few more versions some of the basic flaws that still plague it will be worked out. Till then it's the best I've seen on the market.
Tuesday, October 24, 2006
Linux needs to be less like Windows and more like Mac
Ok, so here's for my latest rant about Linux. I'm sort of over my previous period of maddness and fury having to do with my various linux woes. Since then I have thought a lot about some of the flaws with linux. I have been away from linux for a while because my processor in my desktop had an unfortunate run in with fan failure while I was away and overheating as a result. I have not had the time to tinker with my box, and haven't been home much anyway. So I have been using the computers at work/friend's houses. These are all winxp boxes with a variety of open source applications installed.
There is one exception to this. My boyfriend's room mate has a mac. It is the World or Warcraft box in the dorm room, and as such I spend what little relaxation/game time I have on that computer. Now I have never liked Macs. Going back to the dark ages of computing I always thought Macs were silly, poorly designed and bound to that stupid one button mouse. My opinion of the Mac OS has improved dramatically over the past couple years. The interface is beautiful. I don't like all the eye candy taking up precious system resources, but modern linux distros and windows are both going that direction as well, so I can't complain too much. At least with linux you have the option of picking a distro with slenderness as a design goal. Windows and Mac you're at the whim of whatever you are able to turn off. What has struck me about Mac while using the WoW machine in my boyfriend's dorm room is that if you added a second mouse button (this still irks the ever living innapropriate explitive out of me) it would be a better interface than windows. It's easy to use, responsive, has a logical file system interface as far as I can tell.
Here's the important part. The configuration interfaces are easy to figure out. I mean really really easy to figure out. There is very little you have to tinker with manually, but it is nothing like windows. It doesn't look like windows, things aren't in the same place as windows, the aesthetic is different, the buttons on the top of the program windows are different.
Linux is a nightmare to navigate. If a driver doesn't manually detect, you're probably going to have to go into the text prompt to fix it. Especially if you need ndiswrapper for wifi or something similar. Either it just works or you're going to rip your hair out getting it to work. There is no, well this needs some configuration but the system is standard enough that we could write you a decent graphical interface to do it with. Not with drivers, or for that matter most of the system. Some things aren't like that anymore, samba, cups, x.org. But too much of linux still needs you to go far too deep to get it to work.
Every time someone brings this up the inevitable conversation comes up about linux "becoming like windows". I say we need to be like mac. Not have easy to use config utilities and interfaces in the places Mac has them, but just have them. Make them as easy to use as we possibly can then put them where we think they make sense. Stop worry about "easy" being too like Microsoft. We need to be unique, but start targeting people who are the reason you now have to request a sys restore disk from major OEMs. We are past "Well Linux is easy to install". Who cares if it's easy to install, most people never install an OS, windows, mac, linux or otherwise. The very idea of it terrifys them. We need an interface that Dell would be comfortable putting in a home computer and selling to one of these people. We need a home video program as simple as Mac's, we need a music player as easy and slick as iTunes, and so on and so forth. Before anyone flames me Amarok is great and all, but it looks too much like a program. Audio is one of those consumer electronics things. People want it to looks pretty, and Amarok is just not a pretty program. I'm really hoping songbird fills this void.
Linspire is starting a real honest to god OEM channel program, and god bless them for it. We need more though. Linspire while an ok consumer OS still has K3B as a burning program, and well no offense, but that just doesn't cut the mustard. The wifi support is great, but if you run into anything non standard in a network someone else controlls at a coffeehouse or something you are still out of luck. Xandros is great in some respects, but it looks like you're running something from the Windows 98 Mac OS 8 era, and well that just doesn't cut it either. I don't want to talk about Fedora, Slack, Ubuntu, or any other distros for that matter because quite honestly they are great linux geek distros, and Ubuntu wants to be a distro for the masses, but until it comes with some proprietary sugar coating it's going to be a geek only distro. Like it or not 90% of the population doesn't give two whits about software philosophy and it's time we started admitting that.
There is one exception to this. My boyfriend's room mate has a mac. It is the World or Warcraft box in the dorm room, and as such I spend what little relaxation/game time I have on that computer. Now I have never liked Macs. Going back to the dark ages of computing I always thought Macs were silly, poorly designed and bound to that stupid one button mouse. My opinion of the Mac OS has improved dramatically over the past couple years. The interface is beautiful. I don't like all the eye candy taking up precious system resources, but modern linux distros and windows are both going that direction as well, so I can't complain too much. At least with linux you have the option of picking a distro with slenderness as a design goal. Windows and Mac you're at the whim of whatever you are able to turn off. What has struck me about Mac while using the WoW machine in my boyfriend's dorm room is that if you added a second mouse button (this still irks the ever living innapropriate explitive out of me) it would be a better interface than windows. It's easy to use, responsive, has a logical file system interface as far as I can tell.
Here's the important part. The configuration interfaces are easy to figure out. I mean really really easy to figure out. There is very little you have to tinker with manually, but it is nothing like windows. It doesn't look like windows, things aren't in the same place as windows, the aesthetic is different, the buttons on the top of the program windows are different.
Linux is a nightmare to navigate. If a driver doesn't manually detect, you're probably going to have to go into the text prompt to fix it. Especially if you need ndiswrapper for wifi or something similar. Either it just works or you're going to rip your hair out getting it to work. There is no, well this needs some configuration but the system is standard enough that we could write you a decent graphical interface to do it with. Not with drivers, or for that matter most of the system. Some things aren't like that anymore, samba, cups, x.org. But too much of linux still needs you to go far too deep to get it to work.
Every time someone brings this up the inevitable conversation comes up about linux "becoming like windows". I say we need to be like mac. Not have easy to use config utilities and interfaces in the places Mac has them, but just have them. Make them as easy to use as we possibly can then put them where we think they make sense. Stop worry about "easy" being too like Microsoft. We need to be unique, but start targeting people who are the reason you now have to request a sys restore disk from major OEMs. We are past "Well Linux is easy to install". Who cares if it's easy to install, most people never install an OS, windows, mac, linux or otherwise. The very idea of it terrifys them. We need an interface that Dell would be comfortable putting in a home computer and selling to one of these people. We need a home video program as simple as Mac's, we need a music player as easy and slick as iTunes, and so on and so forth. Before anyone flames me Amarok is great and all, but it looks too much like a program. Audio is one of those consumer electronics things. People want it to looks pretty, and Amarok is just not a pretty program. I'm really hoping songbird fills this void.
Linspire is starting a real honest to god OEM channel program, and god bless them for it. We need more though. Linspire while an ok consumer OS still has K3B as a burning program, and well no offense, but that just doesn't cut the mustard. The wifi support is great, but if you run into anything non standard in a network someone else controlls at a coffeehouse or something you are still out of luck. Xandros is great in some respects, but it looks like you're running something from the Windows 98 Mac OS 8 era, and well that just doesn't cut it either. I don't want to talk about Fedora, Slack, Ubuntu, or any other distros for that matter because quite honestly they are great linux geek distros, and Ubuntu wants to be a distro for the masses, but until it comes with some proprietary sugar coating it's going to be a geek only distro. Like it or not 90% of the population doesn't give two whits about software philosophy and it's time we started admitting that.
Thursday, October 19, 2006
Flash player, development, and the silly FOSS community.
The Flash player beta is live for linux. I am quite pleased that this piece of software is finally available for linux. Though when I saw the blog posting about this going live and I saw the two versions of the player something struck me.
There is a moz plugin and there is a standalone GTK player. I decided to do a little digging and see what dependencies this baby had. What did I find out. I found out that as linux goes it has as close to none as is practically possible. It uses video4linux for video output, it uses basic alsa, X11, GTK, POSIX threads and BSD sockets for non HTTP connections, it also uses openSSL. Now many of these dependencies are for specific features like the video4lin is for webcam input. Most flash doesn't need this. Similarly I imagine computers without some of the later listings would still run the plugin until you tried to use the specific functionality.
This highlights something that really kind of bothers me about how the community has treated Adobe during this process, and it highlights one of the strengths of the closed source development model. This baby will run on just about anything under the sun. I mean really just about anything. Find my a KDE user who is enough of a purist to not have GTK installed on their system. Let's be honest when you think about the history of Qt and Gnome's inception you can venture a guess that they have the lion's share of the purists anyway. So why is this so special, because if this plugin had been developed openly it likely would have used extra development libraries to make things "easier". Would it have gotten out faster, well you bet it would have, and it would have been a beautiful plugin for the people running the distros that properly supported it. Or people running distros that it got into the repositories right away. But even with apt, there have been times, more frequent than I like to list where I have tried to install something and the dependencies weren't in the standard repositories that came listed in my distro, so I had to go in and hunt down new ones that had the required files. This is unacceptable. I am all for distros maintaining quality and perhaps even setting up the version of synaptic on their distro to alert you when you are downloading packages from somewhere other than their main repository, but they should still come connected to a truly comprehensive respository.
The flash plugin avoids all of this, and why do you ask, well likely because the Adobe people reinvented the wheel making the thing. They likely produced all sorts of fun crazy functions and classes that have functionality elsewhere and as a result the program is probably larger than it needs to be, and took a lot longer to get out the door than was necessary, but you know what? I don't care, because it isn't going to be a superior or easier experience on one modern distro than another. It's going to just work, and while I respect the open source development process, and I think it should continue, I wish the FOSS community would start encouraging some closed software where it's appropriate. I like closed and open software. The open software I use, I only use because it is better software. That whole "I don't own a DVD player because it uses CSS" bull that Stallman rants about is in my opinion silly. Open software is a really really good thing. It keeps an open sofware ecosystem, and active software ecosystem, and it keeps the commercial guys on their toes. If it were superior in every way there would be a flash equivalent out. GIMP would be as good as Photoshop, and Blender would be as easy to use as Rhino.
I use GIMP regularly, and I think Blender has come a long way. They are both wonderful programs, but they do not stand up to their commercial counterparts. And the day someone in the open source community takes on Flash I will throw a big ole' here's to the vain hope you keep your sanity party for them. I don't mean vector, I mean a complete open source package integrating vector, and interactivity with a scripting language as dynamic and powerful as actionscript. There isn't even an inferior attempt at this one because no one has been crazy enough to try it.
So give the commercial guys a break, especially when they decide to support us. Maybe if we supported them some the reciprocation would happen more often.
There is a moz plugin and there is a standalone GTK player. I decided to do a little digging and see what dependencies this baby had. What did I find out. I found out that as linux goes it has as close to none as is practically possible. It uses video4linux for video output, it uses basic alsa, X11, GTK, POSIX threads and BSD sockets for non HTTP connections, it also uses openSSL. Now many of these dependencies are for specific features like the video4lin is for webcam input. Most flash doesn't need this. Similarly I imagine computers without some of the later listings would still run the plugin until you tried to use the specific functionality.
This highlights something that really kind of bothers me about how the community has treated Adobe during this process, and it highlights one of the strengths of the closed source development model. This baby will run on just about anything under the sun. I mean really just about anything. Find my a KDE user who is enough of a purist to not have GTK installed on their system. Let's be honest when you think about the history of Qt and Gnome's inception you can venture a guess that they have the lion's share of the purists anyway. So why is this so special, because if this plugin had been developed openly it likely would have used extra development libraries to make things "easier". Would it have gotten out faster, well you bet it would have, and it would have been a beautiful plugin for the people running the distros that properly supported it. Or people running distros that it got into the repositories right away. But even with apt, there have been times, more frequent than I like to list where I have tried to install something and the dependencies weren't in the standard repositories that came listed in my distro, so I had to go in and hunt down new ones that had the required files. This is unacceptable. I am all for distros maintaining quality and perhaps even setting up the version of synaptic on their distro to alert you when you are downloading packages from somewhere other than their main repository, but they should still come connected to a truly comprehensive respository.
The flash plugin avoids all of this, and why do you ask, well likely because the Adobe people reinvented the wheel making the thing. They likely produced all sorts of fun crazy functions and classes that have functionality elsewhere and as a result the program is probably larger than it needs to be, and took a lot longer to get out the door than was necessary, but you know what? I don't care, because it isn't going to be a superior or easier experience on one modern distro than another. It's going to just work, and while I respect the open source development process, and I think it should continue, I wish the FOSS community would start encouraging some closed software where it's appropriate. I like closed and open software. The open software I use, I only use because it is better software. That whole "I don't own a DVD player because it uses CSS" bull that Stallman rants about is in my opinion silly. Open software is a really really good thing. It keeps an open sofware ecosystem, and active software ecosystem, and it keeps the commercial guys on their toes. If it were superior in every way there would be a flash equivalent out. GIMP would be as good as Photoshop, and Blender would be as easy to use as Rhino.
I use GIMP regularly, and I think Blender has come a long way. They are both wonderful programs, but they do not stand up to their commercial counterparts. And the day someone in the open source community takes on Flash I will throw a big ole' here's to the vain hope you keep your sanity party for them. I don't mean vector, I mean a complete open source package integrating vector, and interactivity with a scripting language as dynamic and powerful as actionscript. There isn't even an inferior attempt at this one because no one has been crazy enough to try it.
So give the commercial guys a break, especially when they decide to support us. Maybe if we supported them some the reciprocation would happen more often.
Wednesday, October 11, 2006
Review of Google Docs
So just like everyone else I saw that Google had released this new Google Docs thing. So I thought I would take a look at the program to see what it was all about, and I have to say it's pretty exciting. It isn't going to replace OpenOffice for doing serious page layout. It is rudimentary in a way that I don't even remember Abiword being back when I first used it in 2000. The interface is slow and awkward, and I don't have even a tenth of the options I use in OpenOffice, which given how I rant about word processing programs having too many features for what they are supposed to do, that is saying something. However, it provides all the basic necessities, and allows me to save the documents on a server where I can get to them from anywhere. I don't have to save them on my USB key and use up precious writes to my poor flash memory. I don't have to burn them off to a CD or e-mail them to myself, or upload them to an FTP. I can just save them directly onto storage, and I can choose other people to have access to them, both plain viewing access as well as collaborative access. In fact there is a little line of text in the bottom of the screen telling me no one else is currently editing this very document and a link to Add Collaborators. This program in many ways reminds me of songbird. It lacks a lot of things that everyone else on the market has, but it brings a lot of engaging new functionality to the table that is far more exciting that getting all the old stuff right again for the who cares numberth of times.
I think one of two things will happen with Google Docs. Either google with aggressively develop and market this delightful little morsel of code and it will develop a significant niche, or perhaps grow into a full fledged productivity and collaboration product, or the features available in this program will over the next few production cycles begin making their way into the established products with the old reliable and oh so unfortunately necessery functions established and polished.
The biggest possibilities for this program lie in it's ability to interface with Blogger. The possibilities available in collaborative blogging are delightful, and while Google Docs lacks much of the functionality that I have come to love so much from my mainstay office productivity programs, as a blog publishing program it is far more advanced than anything I've seen in a wordpress interface, or a blogger interface before. By positioning this technology in such a way that it is compared to established online production and publication technologies instead of office productivity technologies Google has given it a market fertile with changes for development, and who knows it may well develop into something that could eventually displace or at least play with the bigger static boys. So I would say take a look, and think about what you do on a daily basis, you might just find yourself using this little AJAX wonder more than you thought.
Thursday, August 17, 2006
Why do podcasts have a freakin XML extension.
Now, I know that podcasts are written in xml. But really, does that actually matter. How hard would it have been in the specification to have the file extension for the xml file to be oh I don't know. . . .pod, or .pdc, or .pd, or anything other than .xml?
Why does this matter do you ask? It matters because I'm working for a radio station and no one knows what to do with the podcast link. They say "what is all this gobledy gook code when I click on the podcast hyperlink? If there were a unique extension even if it was still holding a standard text file with all standard XML code in it, you could bloody well associate it with Itunes, or songbird, or amarok, or any number of other audio players that support podcasts. This seems like a major oversight, and I haven't seen anyone deal with it yet. Tell me this is up for some sort of rewrite in the next version of the podcast spec. Oh wait, probably not.
Why does this matter do you ask? It matters because I'm working for a radio station and no one knows what to do with the podcast link. They say "what is all this gobledy gook code when I click on the podcast hyperlink? If there were a unique extension even if it was still holding a standard text file with all standard XML code in it, you could bloody well associate it with Itunes, or songbird, or amarok, or any number of other audio players that support podcasts. This seems like a major oversight, and I haven't seen anyone deal with it yet. Tell me this is up for some sort of rewrite in the next version of the podcast spec. Oh wait, probably not.
Monday, August 14, 2006
Linux and my spleen on a stick.
Ok, now I know no one will ever read this because who cares about my little bloc, but I'm going to rant anyway.
LINUX SUCKS!!!!!!!!!
Ok there I have gotten that out of my system. Now I don't really believe this, it's just the intense intoxication of the past few days talking. Intoxication brought on by nothing working in either of the Linux environments I was working in.
So first I download freespire. I have been chomping at the bit for the final release of this baby. I'm sick and tired of everyone moaning and complaining about the fact that freespire is including proprietary software with their distro. Cry me a river. I understand they had to get licenses, and I understand Ubuntu and Fedora and everyone else wanting to play it safe legally. I run Kubuntu normally (though I will get to why I am fed up with them soon). And up until recently have been very satisfied with the experience, but seriously if Linspire wants to put out a different product don't rip them for it, just don't use it. It doesn't serve the geek segment. It isn't targeted at the geek segment, and they even made an all OSS version for the geek segment. So stop thrusting your silly little idealigical paradigm on all the non geeks. They end up resenting us for it.
So I boot up the freespire live CD to find out of it supports my particular wireless card and I am greeting with a list of the wireless networks in my area. YAY!!! The drivers work, I won't have to install windows on my linux box when my boyfriend goes back to the dorm. I can use my apartment complex's wireless networks. . . . Oh wait never mind I can't get an IP. So I search the freespire forums. I see tons of other wifi problems (this makes me oh so optimistic about what it's going to be like if I actually fix my current problem). I sadly find nothing on my problem. So I do a general linux wifi search and find one problem that looks like mine and it says that when I leave eth0 enabled sometimes it messes with other network connections. Now this is just brilliant. Freespire has a profile manager to deal with this sort of thing, but say I don't like freespire and I take the time to make ndiswrapper work and I try to be able to accomodate my wireless network at home, and the various wireless networks at Starbucks, Bloomington Bagel Company, the Airport, my friend's house, random hotels, and occasionally accomodate the random hotel networks that require you to plug into an ethernet jack in the room. I was in such a hotel just a couple months ago and this was at the Palmer House Hilton in Chicago, so they aren't that obscure. This is a pretty big problem if I don't want to spend all my time enabling and disabling random network interfaces. Which for the record I don't.
So I disable eth0 and oh it still doesn't work. Bah. I decide I'm sick of it take the liveCD out of my boyfriend's computer and let it boot back into windows where I know my wireless will work I'll just have to deal with everything running at a third the speed it should because Microsoft hasn't cared about OS efficiency and productivity since . . . oh wait nevermind.
So now I jump back to Kubuntu where I boot up with a new mouse. It has a scroll wheel which my last on didn't, and it's a standard ps2 mouse. Kubuntu now looks like satan has taken over as random input from the ps2 interface completely overwhelms the KDE interface and random programs start and stop in a constant dance of insanity and hate. Time to try that freespire liveCD again as I need to burn some files off of my ReiserFS hard drive. So I boot into Freespire and my mouse is mostly fine. There is a little twitch to it that makes me think maybe the hardware is a little flakey, but when I say twitch I mean the cursor shutters a couple millimeters every now and then and then goes back to normal. This is hardly the manifestation of Beelzabub that showed up on my screen in Kubuntu. So I go to burn my anime so I can be distracted at work during the slow periods an activity which my employer is perfectly kosher with. The first 4 episodes I need burn slowly but they burn. The next three will not. Now my CD burner drive is old and has personality so this I will not blame on linux, it is totally not the OS's fault. It however leads me to another interface abomination that really is the OS's fault. I throw in my USB key, so I can grab at least one more episode to bring with. It won't fit. I get 95 percent and boom it's out of space. Now does it tell me it's out of space? No, the file transfer stalls and nothing more happens. I'm sorry, I don't care what weird wacked out world you live in. No OS, not even windows 3.0 has any excuse for not being able to tell when you've maxed out a supported file system. So I cancel the file transfer. Go in and shred the partial file. I'm a touch peeved that Freespire didn't do this for me when I canceled the file transfer. Again this is windows 3.0 functionality here. Come on people. Then I delete a couple other things so I can get my Loki fix, and I try again. I only get about 6 megs transfered. NANI?!?!?
After one more attempt where I get 3 megs transfered I take a look at the available space on my key, and there is none. I shredded the files. It tried to put them in the trash can but I stopped that right away because maintaining a trash can for a USB disk is sort of idiotic. Post shredding Linux didn't update the file system. Now I know Unix saves deletes and file reorganization for when the system isn't busy or for shutdown. It just logs that it needs to do it. I remember from back in my RedHat 6.0 days that this is a big part of why you never want to just kill your computer. Proper shutdown is a must. This is a server mentality. Servers have lots of resources to work with in the way of disk space, so a little virtual linking dosn't hurt anyone, and the performance hit of actually altering a file system while the computer is being actively used makes a difference on a server. This is a desktop OS. No one in their right mind is going to say Freespire is designed for server work, and it's a freaking mounted USB thumb drive. I mean come on. When I say shred the file shred the damned file. I shutdown stuck my thumb drive in my boy's windows comp to make sure things weren't damaged and everything was fine. What if I had taken the thumb drive out before shuting down though? As this would be the normal situation. What if?
I could go on about my complaints over the past month with at least 3 other distros including DMS, Puppy Linux, Vector Linux, and Xubuntu. All these were attempt to make my boyfriends old 500 Mhz 60 Megs of Ram laptop usable again. Which I finally pulled off with Vector linux, but only after working around one of the more idotic bugs ever.
Linux is fine for tinkering, it's great for servers, but until these really basic interface stupidites are worked out it's never going to be Year of Linux on the Desktop. Are there ways of working around these problems. Sure there are. There are always ways of working around problems. But the fact of the matter is why is the averag user going to want to put in that extra effort when they don't have to with windows. Will linux run faster, sure it will, but you know what unless you're doing a bunch of multi tasking and high profile computing which means you are more likely to be a linux or Mac person to begin with then who cares. The computer will run fast enough, even if it's only a $299 e-machine. I expected more from the community distro run by Linspire, which is supposed to be one of the linux for the masses distros. If they can't get stuff this simple right then we might as well lay down for Redmond now.
LINUX SUCKS!!!!!!!!!
Ok there I have gotten that out of my system. Now I don't really believe this, it's just the intense intoxication of the past few days talking. Intoxication brought on by nothing working in either of the Linux environments I was working in.
So first I download freespire. I have been chomping at the bit for the final release of this baby. I'm sick and tired of everyone moaning and complaining about the fact that freespire is including proprietary software with their distro. Cry me a river. I understand they had to get licenses, and I understand Ubuntu and Fedora and everyone else wanting to play it safe legally. I run Kubuntu normally (though I will get to why I am fed up with them soon). And up until recently have been very satisfied with the experience, but seriously if Linspire wants to put out a different product don't rip them for it, just don't use it. It doesn't serve the geek segment. It isn't targeted at the geek segment, and they even made an all OSS version for the geek segment. So stop thrusting your silly little idealigical paradigm on all the non geeks. They end up resenting us for it.
So I boot up the freespire live CD to find out of it supports my particular wireless card and I am greeting with a list of the wireless networks in my area. YAY!!! The drivers work, I won't have to install windows on my linux box when my boyfriend goes back to the dorm. I can use my apartment complex's wireless networks. . . . Oh wait never mind I can't get an IP. So I search the freespire forums. I see tons of other wifi problems (this makes me oh so optimistic about what it's going to be like if I actually fix my current problem). I sadly find nothing on my problem. So I do a general linux wifi search and find one problem that looks like mine and it says that when I leave eth0 enabled sometimes it messes with other network connections. Now this is just brilliant. Freespire has a profile manager to deal with this sort of thing, but say I don't like freespire and I take the time to make ndiswrapper work and I try to be able to accomodate my wireless network at home, and the various wireless networks at Starbucks, Bloomington Bagel Company, the Airport, my friend's house, random hotels, and occasionally accomodate the random hotel networks that require you to plug into an ethernet jack in the room. I was in such a hotel just a couple months ago and this was at the Palmer House Hilton in Chicago, so they aren't that obscure. This is a pretty big problem if I don't want to spend all my time enabling and disabling random network interfaces. Which for the record I don't.
So I disable eth0 and oh it still doesn't work. Bah. I decide I'm sick of it take the liveCD out of my boyfriend's computer and let it boot back into windows where I know my wireless will work I'll just have to deal with everything running at a third the speed it should because Microsoft hasn't cared about OS efficiency and productivity since . . . oh wait nevermind.
So now I jump back to Kubuntu where I boot up with a new mouse. It has a scroll wheel which my last on didn't, and it's a standard ps2 mouse. Kubuntu now looks like satan has taken over as random input from the ps2 interface completely overwhelms the KDE interface and random programs start and stop in a constant dance of insanity and hate. Time to try that freespire liveCD again as I need to burn some files off of my ReiserFS hard drive. So I boot into Freespire and my mouse is mostly fine. There is a little twitch to it that makes me think maybe the hardware is a little flakey, but when I say twitch I mean the cursor shutters a couple millimeters every now and then and then goes back to normal. This is hardly the manifestation of Beelzabub that showed up on my screen in Kubuntu. So I go to burn my anime so I can be distracted at work during the slow periods an activity which my employer is perfectly kosher with. The first 4 episodes I need burn slowly but they burn. The next three will not. Now my CD burner drive is old and has personality so this I will not blame on linux, it is totally not the OS's fault. It however leads me to another interface abomination that really is the OS's fault. I throw in my USB key, so I can grab at least one more episode to bring with. It won't fit. I get 95 percent and boom it's out of space. Now does it tell me it's out of space? No, the file transfer stalls and nothing more happens. I'm sorry, I don't care what weird wacked out world you live in. No OS, not even windows 3.0 has any excuse for not being able to tell when you've maxed out a supported file system. So I cancel the file transfer. Go in and shred the partial file. I'm a touch peeved that Freespire didn't do this for me when I canceled the file transfer. Again this is windows 3.0 functionality here. Come on people. Then I delete a couple other things so I can get my Loki fix, and I try again. I only get about 6 megs transfered. NANI?!?!?
After one more attempt where I get 3 megs transfered I take a look at the available space on my key, and there is none. I shredded the files. It tried to put them in the trash can but I stopped that right away because maintaining a trash can for a USB disk is sort of idiotic. Post shredding Linux didn't update the file system. Now I know Unix saves deletes and file reorganization for when the system isn't busy or for shutdown. It just logs that it needs to do it. I remember from back in my RedHat 6.0 days that this is a big part of why you never want to just kill your computer. Proper shutdown is a must. This is a server mentality. Servers have lots of resources to work with in the way of disk space, so a little virtual linking dosn't hurt anyone, and the performance hit of actually altering a file system while the computer is being actively used makes a difference on a server. This is a desktop OS. No one in their right mind is going to say Freespire is designed for server work, and it's a freaking mounted USB thumb drive. I mean come on. When I say shred the file shred the damned file. I shutdown stuck my thumb drive in my boy's windows comp to make sure things weren't damaged and everything was fine. What if I had taken the thumb drive out before shuting down though? As this would be the normal situation. What if?
I could go on about my complaints over the past month with at least 3 other distros including DMS, Puppy Linux, Vector Linux, and Xubuntu. All these were attempt to make my boyfriends old 500 Mhz 60 Megs of Ram laptop usable again. Which I finally pulled off with Vector linux, but only after working around one of the more idotic bugs ever.
Linux is fine for tinkering, it's great for servers, but until these really basic interface stupidites are worked out it's never going to be Year of Linux on the Desktop. Are there ways of working around these problems. Sure there are. There are always ways of working around problems. But the fact of the matter is why is the averag user going to want to put in that extra effort when they don't have to with windows. Will linux run faster, sure it will, but you know what unless you're doing a bunch of multi tasking and high profile computing which means you are more likely to be a linux or Mac person to begin with then who cares. The computer will run fast enough, even if it's only a $299 e-machine. I expected more from the community distro run by Linspire, which is supposed to be one of the linux for the masses distros. If they can't get stuff this simple right then we might as well lay down for Redmond now.
Monday, June 19, 2006
Why do people hate Flock???
Ok I have to rant about a couple technological things. They are not directly related to Flock specifically, but they have been active in the Flock world recently. At least the world of "users". Why do so many open source people have a problem with Flock? I can't tell you how many times I've seen people complain that "Oh Flock is just Firefox with some plugins". Now if you just look at a feature list this is technically true, but if you have used Flock and Firefox with said plugins then you know what a monumental difference there is. What's lacking from Firefox with the plugins. Drumroll please. . . . ."USABILITY RESEARCH!!!!!"
After reading the other day about the Better Desktop Initiative it's obvious to me that the people getting paid to develop open source software understand that software by programers for programers will never go anywhere in the general market. Let me say this again. Software by programers for programers is USELESS to non-programers. And most firefox plugins are software by programers for programers. Not all of them are, many of them are wonderful, Forecastfox is fabulous, as are the majority of the transparent plugins that are meant to speed up the performance of your browser or make pdf downloading easier, or other such niceties. However, the RSS reader options in the firefox plugin section for lack of a better description are terrible, and I haven't found any decent blogging options. Why? Because they lack integration with each other. Do the blogging plugins work with the photo upload plugins, work with the shared bookmark plugins? No they don't. And what would it take for these plugins to work together. A unified effort with unified developers working on all of them, and someone getting paid in some manner so the project stayed alive. Hmmm. Now in order for this to happen one would assume that one would need a product. Something that you could eventually sell, or at least give away and have it bring people to a sponsored search partner or some advertising media to bring in revenue.
Now I want all you developers out there to stop being a developer for a minute. Now think about advertising. Try to get inside the head of a person who doesn't know what a plugin is, or how to configure one, or has the expertise to navigate an interface with no formal usability testing behind it. Now will this person respond to unified plugins (translation, ew I'm gonig to have to configure it myself), or a social networking browser that brings together all your favorite features for you? (translation Yay! someone else got paid to configure it for me!!!).
I am so sick of hearing about "Linux on the Desktop" and "Of course linux is good enough for everyone" and "why are you still using office, when Open Office will do everything you need, and it's so similar in interface (anyone who's ever watched a non techie go from office to openoffice and try and find the word count knows there are enough differences to matter to the base consumer, even if not to the basic adept corporate receptionist).
For the record. I don't use Flock, I've downloaded every version to see what it's about, and I find it exciting. I am a techie. I don't care about it personally, and I run a linux desktop/server, and I have used OpenOffice since it was StarOffice 5.1, and I used several of the OpenOffice pre 1.0 releases. Going back to 2000. I can do all of that, but I also know most people can't. The year of the Linux Desktop will come when the people giving back to open source projects for free understand they aren't coding for their buddies. They are coding for the guy down the street that coaches basketball, and his wife who has taught French since 1985. We all know their kids can use our stuff, and yes if they got over they psychological block against change in technology we all know they are intelligent to use our stuff as well. But they don't want to get over that block, it isn't worth it to them, and it's time we stopped asking them to.
Novel/RedHat/Xandros are not going to have the kind of money for usability research that Microsoft has any time soon, and Microsoft already has a huge head start on them and a larger paid developer base than they do to keep that lead. Our advantage is our numbers, but the numbers need to know what they are facing and honestly approach it. And for the record Kudos to those few of you who do.
After reading the other day about the Better Desktop Initiative it's obvious to me that the people getting paid to develop open source software understand that software by programers for programers will never go anywhere in the general market. Let me say this again. Software by programers for programers is USELESS to non-programers. And most firefox plugins are software by programers for programers. Not all of them are, many of them are wonderful, Forecastfox is fabulous, as are the majority of the transparent plugins that are meant to speed up the performance of your browser or make pdf downloading easier, or other such niceties. However, the RSS reader options in the firefox plugin section for lack of a better description are terrible, and I haven't found any decent blogging options. Why? Because they lack integration with each other. Do the blogging plugins work with the photo upload plugins, work with the shared bookmark plugins? No they don't. And what would it take for these plugins to work together. A unified effort with unified developers working on all of them, and someone getting paid in some manner so the project stayed alive. Hmmm. Now in order for this to happen one would assume that one would need a product. Something that you could eventually sell, or at least give away and have it bring people to a sponsored search partner or some advertising media to bring in revenue.
Now I want all you developers out there to stop being a developer for a minute. Now think about advertising. Try to get inside the head of a person who doesn't know what a plugin is, or how to configure one, or has the expertise to navigate an interface with no formal usability testing behind it. Now will this person respond to unified plugins (translation, ew I'm gonig to have to configure it myself), or a social networking browser that brings together all your favorite features for you? (translation Yay! someone else got paid to configure it for me!!!).
I am so sick of hearing about "Linux on the Desktop" and "Of course linux is good enough for everyone" and "why are you still using office, when Open Office will do everything you need, and it's so similar in interface (anyone who's ever watched a non techie go from office to openoffice and try and find the word count knows there are enough differences to matter to the base consumer, even if not to the basic adept corporate receptionist).
For the record. I don't use Flock, I've downloaded every version to see what it's about, and I find it exciting. I am a techie. I don't care about it personally, and I run a linux desktop/server, and I have used OpenOffice since it was StarOffice 5.1, and I used several of the OpenOffice pre 1.0 releases. Going back to 2000. I can do all of that, but I also know most people can't. The year of the Linux Desktop will come when the people giving back to open source projects for free understand they aren't coding for their buddies. They are coding for the guy down the street that coaches basketball, and his wife who has taught French since 1985. We all know their kids can use our stuff, and yes if they got over they psychological block against change in technology we all know they are intelligent to use our stuff as well. But they don't want to get over that block, it isn't worth it to them, and it's time we stopped asking them to.
Novel/RedHat/Xandros are not going to have the kind of money for usability research that Microsoft has any time soon, and Microsoft already has a huge head start on them and a larger paid developer base than they do to keep that lead. Our advantage is our numbers, but the numbers need to know what they are facing and honestly approach it. And for the record Kudos to those few of you who do.
Tuesday, June 13, 2006
Testing the Flock Beta
I don't think I've ever seen the basic interface of a program change so much between 0.2 worth of version increment. Flock has grown up a lot, and lost a little bit of it's early energetic charm I unfortunately have to say, but I think it may well be a more mature and hopefully usable project. We'll see, for now I just want to see how well it updates.
Subscribe to:
Posts (Atom)