Modify Colors
Posts in Opinion
Illustrating the blessing and the curse of open-source software, some contriversy has arisen over Intel’s Moblin (their Linux for netbooks) replacing NetworkManager with a custom program called ConnMan.
NetworkManager is tool for configuring network connections, from ethernet to 3G. NetworkManager is used in almost every modern Linux distribution, pretty much regardless of the desktop environment. Due to complaints about NetworkManager’s architecture, extensibility, customization options, and lack of WiMax support, the Moblin project decided they needed something completley new: ConnMan.
ConnMan lacks many of the features of NetworkManager, but offers complete separation of the front-end and back-end, WiMax support, and more customization/extensibility.
The Blessing
Open-source software allows anyone to have the choice of rewriting a piece of something that they don’t think is done the right way or that just isn’t right for them. This means that when one project starts sliding in quality, another project can create an alternative program to do the same, or a similar, thing, often utilizing some of the existing code. This often prevents applications from being abandoned and becoming obsolete.
The Curse
This same attribute of open-source software is also a curse, though. A lot of time is spent rewriting code that, in some cases, may be perfectly good code. This time could be better spent working on something new. Worse, when you get two conflicting applications, as is the case here, it fragments the Linux desktop. Right now, every major desktop Linux distribution is using NetworkManager. This means that when one distribution improves the tool, everyone benefits. If, however, a few of these distributions switched to ConnMan, suddenly only the ConnMan distros would benefit from other ConnMan distros and the same for NetworkManager distros.
This is simply an attribute of open-source. Sometimes it is incredibly beneficial, such as when a project is abandoned, but it also encourages people to reinvent the wheel over and over again.
Don't Force Purity in Companies
A recent ComptuerWorld article argues that open-source-based companies are just not going to be able to make as much money as their commercial counterparts, and, thus, should open themselves up to supporting the open-source versions of their software, too. I disagree.
These companies are already providing an incredible service to the community by licensing their code as open-source. True, they may get help from the community in developing the software, but they still need a way to make money. Compared to some of the alternatives, only supporting the commercial version doesn’t seem too bad, and, if it helps the company hire more developers to improve the open-source side, everyone wins.
Though it would be nice if every company was based around pure open-source, it may not yet be realistic, at least until someone comes up with a better business model. For now, let a few concessions slide – you can’t ask too much of company that has to make money.
Not at all surprisingly, I received a lot of negative feedback about this recommendation. Still, I stand by it. Here is why:
A new user using Ubuntu for the first time as their main operating system is bound to run into some snags. After all, whenever you switch to a different operating system there are almost certain to be some differences that snag you. While many of these differences are just necessary, in situations where they can be limited and made easier, that should be done. Having Open Office save in .doc is one of those situations where a new user’s life can be made much easier, simply by having Open Office save in a format they, and their co-workers, are used to.
The most frequent counter argument to this is that it is better to encourage new users to use ODF, because it is an open standard. While I recognize the importance of this, I think it is even more important not to scare off new users. The idea that sending someone an ODF file will make then go download Open Office and suddenly transform them into a happy convert is not likely to happen. More likely, that person will ask you to send them a readable format. Even if no other person is involved, what if you need to edit your files on a computer with only Office. Of course you could install Open Office, but a new user might not know to do this. Worse, you could be at a computer where you do not have sufficient rights to install an application.
I believe that it is far better to allow a new user to make the choice to use open standards on their own, rather than forcing it on them and potentially making them run away from everything open.
The iPhone Is Now Doomed
I cannot claim to be a developer, but I have been watching the whole iPhone application development issues with interest. As of today’s news, it appears that the iPhone development process is like this:
- Ask Apple for permission to make an application.
- Sign a non-disclosure agreement.
- Invest time and money into an iPhone application.
- Ask Apple for permission to sell or give away your application.
- If Apple says YES: start making money and hope Apple does not change their minds.
If Apple says NO: shut up and deal with it. If you say anything, Apple can sue you, further raising the wasted investment money.
It just makes no sense.
For months now, Apple, the company known for being good at generating a lot of good free PR, has received an almost continuous stream of negative press, first over MobileMe, then the 2.0 software, then over iPhone applications being semi-arbitrarily not approved, and now Apple has extended the NDA so developers cannot say why their applications were not approved. Essentially, Apple has responded to push back over a too tightly controlled system by controlling it even tighter. Worse, Apple managed to time this perfectly with the release of Android, a completely open platform.
Apple has not shot itself in the foot. They shot themselves in the leg or heart. If Apple does not loosen up on their NDA policies soon, developers may leave the iPhone for the much more open Android platform or another more open platform. If the developers leave, Apple has suddenly doomed a potentially promising and incredibly successful platform. The only question that remains to be seen is how far iPhone developers are willing to be pushed? My guess: not much more.
The Theory Of Google's Chrome As An Operating System
There is no doubt that Google did a nice job of promoting their “Chrome” web browser. Practically everyone is considering this front page news, and for a good reason, too. Whatever you think of it, Google, primarily a web search and web application company, introducing a web browser is big news. The interesting part of all the coverage is the theories about why Google made this move.
Many of the theories around propose that Chrome is really the long-awaited “Google OS” in the form of a platform for web applications. A couple of days ago, I wrote my own theory, which proposed that Chrome is really about pushing other web browsers forward to better support web applications in general. Both of these theories are really quite similar, in that they are both making it easier for people to use web applications to replace desktop applications.
The real difference is that one theory suggests that Google’s Chrome will become the platform for web applications, meaning that it will have specific, unique features that will benefit Google applications, and perhaps nothing else. In the other theory, my theory, Google would use Chrome to push all browsers along, something which would be beneficial to all. Only time will tell what their plans are (after all, they may not even know yet), but I want to give a warning if Google follows the path of creating their own custom platform: Avoid it. Just avoid it.
Why? The web is a chance to redefine applications from a piece of software that runs on a particular OS to just a piece of software. Right now, almost by definition, web applications are cross platform. This is our chance to keep it like that. If Google introduces their own custom platform for their web applications, that is a major step in the wrong direction. It is almost guaranteed to end up with a mess of “features” introduced by Google to be used by Google. We have to keep the platform maker separate and independent from the software maker if we want to maintain the standards of the web. No single group should even be able to dominate both fields. If they do, consumers lose.
All of this is not to say that Chrome is evil. If the point, as I suspect it is, is just to push web browsers into the future, then that is great. We just have to keep a watchful eye and not let Google get too much power.
Shuttleworth Is Right – Apple is Linux's Main Competition
A lot of people view Windows as Linux’s main competition, largely because Microsoft has, by far, the majority of the market and is considered “the standard” by most people. Mark Shuttleworth and I disagree.
The Var Guy, and plenty of other people, are reporting that Mark Shuttleworth, founder of Canonical, said during his keynote at OSCON, that Linux must not just match but exceed Apple’s Mac OS X, strongly implying that Apple – not Microsoft – is Linux’s main competitor. I agree. Here is why:
First, Microsoft has not shown that is headed anywhere except down. Net Applications shows Windows taking a steady fall since 2006, which appears to be as far back as their data goes. Unless Microsoft really gets itself together, there is no reason why this should change.
Second, I honestly cannot come up with many arguments for Windows, but I can come up with some for the Mac, even if I don’t agree with them. Seriously, why choose Windows over Linux?
- It is the standard and everything is compatible with it.
- You prefer Windows’s way of installing software…?
I am sure I am missing something, but I really cannot figure out what. On the other hand, why choose the Mac over Linux?
- The closed system. Personally, I don’t think that is an advantage, but you could make the argument that things work better because Apple controls so much of the ecosystem. Let me put it this way, no one has proved that either the open approach or the closed approach is the “right” way.
- You prefer the Mac’s way of installing software…?
- You like the Mac interface better than any of the other options.
- I would guess that more (edit: I should add “external” here.) devices work under Mac OS X than under Linux.
Once again, I am sure I am missing some arguments here, but my main point is that, as a technical user, I can see the argument for the Mac, but the argument for Windows seems… weak.
Overall, it looks to me, and some would very much disagree with this, like the future is not going to be Windows, so why fight it?
As Ubuntu Intrepid Ibex development continues, Phoronix is reporting that there has been some discussion on the ubuntu-devel mailing list about removing or disabling bullet-proof X, a feature designed to make it so that a user is never kicked into a command line interface (CLI) due to display problems. The basic argument for doing this is that there have been a number of unfixed bugs in bullet-proof X and, as pointed out by Phoronix, advanced users may prefer being kicked to a CLI to debug the problem.
While these are reasonable arguments, I don’t think dropping a good idea, in this case bullet-proof X, is, well, a good idea. To be fair, there is some discussion of replacing bullet-proof X with a more integrated solution to accomplish the same thing, which would be fine. What I am worried about is simply dropping it and potentially scaring new users away.
Right now nothing is certain, though, so don’t get too upset.
As you know, I am absolutely against making the same hardware cost the same with two different operating systems that don’t cost the same price. So if you take a given PC or laptop, put a free OS on one version and a commercial OS on the other and charge the same price, that is unfair. You may wonder why I am making such a big deal of the same hardware. Obviously you can’t compare across two different PCs, right? Well, apparently not.
A recent Techworld article is titled “Windows same price as Linux in new Eee PC.” When I read this, I thought “Ugg. Asus is messing with prices again.” Then I read the article.
If you look closely, the end of the second sentence says “although the specs are different.” That makes the story completely different and makes the title very misleading.
Just to make sure I am very clear, there is nothing wrong with selling machines with different priced OSs, if there is a difference in the specs that compensates for the price. Please, no more misleading titles.
Would More Expensive Linux PCs At Wal-Mart Work?
In a recent Datamation article, Matt Hartley put forward two suggestions for making future Linux PCs a success in stores like Wal-Mart. As many of you may know, Wal-Mart’s experiment with selling the gPC, a cheap Linux-based PC, was essentially a failure. I certainly agree with Hartely’s second suggestion for making future “Linux in Wal-Mart” experiments more of a success, which is to give consumers enough information to be sure that everything they use will work with Linux. Unfortunately, I am not as sure I agree with his first suggestion.
Hartely’s first suggestion is that Linux PCs sold should not use such low-end hardware. At first this makes sense, but if you add higher end hardware, the PC gets more expensive. The problem with this is that the $50-ish price saving you get from Linux is much less relavent if the PC costs $1000 than if it costs $200. That might be fine if someone was looking for Linux, but most Wal-Mart shoppers are probably just looking for a good deal. Perhaps a better solution than adding more expensive hardware is to use a less resource-intensive Linux distro such as Xubuntu or DSL. It would certainly be interesting to see a stripped down Linux compete with Windows Vista.
Also, if you are interested in a live group distro review, please vote in the poll by clicking on this link.
Linux's Chance to Leap Ahead of Apple And Microsoft
Apple and Microsoft are both in a situation where they are essentially forced, due to past mistakes or a lack of foresight, to spend a significant amount of time making behind-the-scenes changes. In this time, Linux could potentially get a real leg up on other OSs.
First, what happened with Microsoft and Apple? In Microsoft’s case, the main mistake was making Vista so resource intensive. When the Eee PC suddenly appeared as a major threat to Windows, Microsoft was not prepared. As a result, Microsoft has had to repeatedly extend the life of Windows XP, a product they would obviously rather retire. This mistake, making Vista so resource intensive, will presumably force them to make major changes in Windows 7 to bring down the resource requirements.
On Apple’s side, OS X 10.6 (Snow Leopard) has just been announced. Rather than introducing new features, though, Apple has decided to essentially do a maintenance release. I was very surprised by this decision, but apparently that is the plan.
So what will Linux be doing well Apple and Microsoft are essentially standing still? The short answer, I hope, is: running. There are two reasons why Linux will not have to take time off for a “maintenence release.” First, Linux has not made mistakes such as dramatically increasing resource requierments. Second, even if some people (even an entire distribution) decide to work on only behind-the-scenes stuff, new features will continue to flow in from projects that are moving ahead. No one project or person can stop Linux innovation. Because of this, I am hopeful that Linux will be able to push ahead of Microsoft and Apple as they stand there correcting past mistakes. Anyone thinking of the tortoise and the hare?
Also, please check out the information about the live, interactive, group distro review.