Practical Technology

for practical people.

May 10, 2010
by sjvn01
2 Comments

Sun Open Source Reborn in ForgeRock

When Oracle bought Sun, there were many unanswered questions about Sun’s open-source portfolio of programs. Over a year later, we still don’t know, for example, if OpenSolaris is going to have Oracle’s support. We now know, however, that OpenSSO, an open source access management and federation server platform, will live on as a product under the new open-source company ForgeRock.

ForgeRock is an ISV (independent software vendor) made up of former Sun business and technology experts. The company claims that it based its “business exclusively on open source products.” The first of these is its I3 Open Platform. This identify management suite is built on top of several open-source including OpenAM, which is based on OpenSSO); OpenESB; OpenIdM; and OpenPortal, which is built on LifeRay.

According to Simon Phipps, former chief open source officer at Sun and, as of today, May 10th, a member of ForgeRock’s board and the Norwegian company’s chief strategy officer. ForgeRock’s I3 Open Platform is an open, high-performance and unified platform addressing.

More >

May 7, 2010
by sjvn01
2 Comments

Linux needs to do more for programmers

Much as I hate to admit it, Microsoft does some things better, much better, than Linux. Number one with a bullet is how Microsoft helps programmers and ISVs (independent software vendors). MSDN (Microsoft Software Developer Network) is a wonderful online developer resource. Linux has had nothing to compare.

True, there is the Linux Developer Network, which, when it began, looked like it would be the Linux equivalent of MSDN, but it hasn’t lived up to its promise. And, I can’t overlook the Linux Foundation’s Linux training classes. But, if I’m an ISV and I want to write software for Linux, I’m still going to need to piece together a lot of it by myself.

More >

May 7, 2010
by sjvn01
0 comments

Playing it Safe(r) on Facebook

I’m not ready to give up on Facebook yet, like my friend Jason Perlow, but I can understand why he’s sick to death of it. The simple truth is that Facebook takes lousy care of your personal information and is happy to sell it to advertisers at the drop of a hat.

Now, I lead a pretty public life and I really don’t care if anyone knows, for example, that I like watching Dr. Who; reading the books of Terry Pratchett, or listening to alt.country music. But, I don’t want people ‘listening’ in to my IM (instant message) conversations or Facebook quietly adding applications to my account without me knowing. Facebook is already full of games like Farmville and Mafia Wars that are always trying to get my attention when I have less than zero interest in them. The last thing I need is yet another insecure Facebook App, which I didn’t even ask for.

For now, though, I’m sticking with Facebook. I’m just going to be scaling back a lot of what I’ve made public on my account though. While I’m not going to go as far as Perlow has in his excellent Lockdown or Death for your Facebook Profile, I have tried to make my site a bit more private, and here’s how I went about it.

First, and foremost, when you decide what information you’re going to try to keep private you should keep in mind that Facebook’s security is rotten. If you really don’t want something to be known about you, then don’t just keep it off Facebook, keep it off the Web. If you don’t, you could end up like the woman who lost her job because she had posted about her polyamorous life on Twitter, without realizing that in the brief time she had used her real name on her profile a Twitter search engine had picked it up.

You can argue all you want that she shouldn’t have been fired for what she did outside her job. I’d agree with you. But, the sad, simple truth is that she was fired because her private life wasn’t as private as she thought it was.

Does that scare you enough that you want to delete your Facebook account? Well, as that woman found it may already be too late to hide your deep dark secret. Google, Yahoo, and all the other search engines track down everything public on Facebook just like they do everything else. Even if you decided to delete your Facebook account today, the actual deletion will take at least two weeks and your ‘tracks’ on the search engines will persist for months and even years.

Me? I’m just going to get rid of some information. Yes, an expert can find and drag out most of this stuff from the Web if they know what they’re doing, but that’s the point. Facebook has made it trivial for crackers or advertisers to watch your every move. I just want to make it harder.

That said, head over to Facebook’s Basic Information tab and edit down what you already have on there. Is anyone ever going to need to find your full snail-mail address on Facebook? I doubt it. Out it goes. Do you really want anyone to know how old you are when a potential employer may very well hold it against you that you’re over 50? I don’t think so. Go over your information carefully and unless you can think of a good reason for it to be public knowledge, get rid of it.

Next, you need to be ready to spend a lot of time customizing your Facebook Privacy settings. I’ll make it simple for you though. While Perlow suggests that you use “The most restrictive level [which] is to choose Custom for each of these fields and to set it to Only Me,” I’d suggest that if you feel that way about it you should delete the information in the first place. For everything else, I’d set it to “Only Friends.”

That done, you should pay particular attention to the Privacy Settings Applications and Websites page. This page determines what your friends can share about you. I don’t know about you, but I want to be the one in charge of my information, not my friends. Here, I un-check everything. Facebook, bless their no privacy little hearts, wants to let my friends share everything about me both to Facebook and to other Web sites. I think not.

Another simple, for once, page you should pay particular attention to is the Instant Personalization Pilot Program. This is Facebook’s new toy, which lets Facebook — excuse me, you — share your information with third party Web sites. Maybe you feel the need to share everything that’s public about you on Facebook with, say, Microsoft via Microsoft Docs, but I’m not so trusting. Just say no and don’t allow it access.

The privacy settings let you clean up a lot of the mess, but it doesn’t take care of everything. Let’s say that you want to get out of some groups like, “IAlwaysGetDrunkonFridays.” Facebook doesn’t make it easy to just dump all the silly groups from one page. Instead, you need to go to each group and leave them individually. Look for the “Leave Group” link below the group’s image on its page and click it. Then, move on to the “IalwaysGetDrunkonSaturdays” page and leave it. You get the idea.

Next, let’s get rid of any applications you don’t need. Applications are especially bad about sharing your information with people so you should show no mercy in getting rid of them. To do this, head over to Account, select Application Settings, and then use the “Authorized” filter. Do not, do not, think that you can use the Facebook Applications link to do this. That only shows you a small fraction of all the applications you’ve authorized for access to your account.

If you’re like me, you’ll be amazed at how many applications you’ve given leave to raid your account for information. You could edit each application for how much it can show the world about you, but personally, I found myself zapping almost all of them. The Monty Python Gifts app., for instance, was funny once; it’s not funny for all time.

Once you’ve got all this done, you’ll be relatively safe on Facebook. Of course, Facebook can throw up another security hairball at any time and, goodness knows, we can always make fools of ourselves with our own efforts, but at least this will keep your information as secure as Facebook will currently allow. Good luck.

A version of this story first appeared in ITWorld.

May 5, 2010
by sjvn01
0 comments

Chrome 5: Faster and Better

The first thing you’ll notice with Google’s new beta of its Chrome Web browser is that it’s faster, much faster, than the last version. You don’t need any fancy tests to see that. All you have to do is use it and you’ll see that it blows other browsers away.

But, if numbers are what you want, here’s what I found using the SunSpider JavaScript benchmarks. I ran these tests on a pair of Dell 530S desktop PCs. These older computers are powered by a 2.2GHz Intel Pentium E2200 dual-core processor with an 800MHz front-side bus. Each has 4GB of RAM, a 500GB SATA (Serial ATA) drive, and an Integrated Intel 3100 GMA (Graphics Media Accelerator) chipset. One was loaded with Windows XP SP3 and the other used MEPIS 8.0 desktop Linux. First, I ran the benchmarks with an old copy of Chrome 4, which I installed just for these tests, and then with Chrome 5.0.375.29, the latest beta. The average result was 660.4ms for Chrome 4 and 380.0 for Chrome 5.

Other Chrome tests haven’t shown such significant improvements, but the bottom line is that the new Chrome is visibly faster than most other browsers. While it’s true that Opera Software’s Opera 10.53 is faster still, it’s also true that Opera has been plagued with numerous serious security problems. For the best combination of speed and security you can’t currently beat Chrome.

More >

May 4, 2010
by sjvn01
2 Comments

The other Ubuntu Linux distributions

I like the brand spanking new Ubuntu 10.04 a lot. But while I like its GNOME 2.30 interface, I also like other interfaces such as KDE. It would be nice if Ubuntu could also play MP3s, common video formats and Flash from the get-go. You could install all these and other extras from the Ubuntu repositories, but there’s also a wide-variety of Ubuntu spin-offs that come ready to give you the functionality you want right out of the box.

Here’s my list of the most important of the Ubuntu-based distributions.

Kubuntu 10.04. Like the name suggests, the big difference between Ubuntu and Kubuntu is that the K-Ubuntu runs KDE 4.4.2 instead of GNOME 2.30 for its desktop. But Kubuntu isn’t just Ubuntu with KDE. Instead of KDE’s default Konqueror Web browser, Kubuntu defaults to using Firefox 3.6.3.

Kubuntu also does a nice job of integrating GNOME applications into the KDE 4.4 interface. That’s the good news; the bad news is, instead of using a KDE frontend to Ubuntu’s outstanding Ubuntu Software Center, you’re stuck with the far less attractive and more difficult to use KPackageKit to add new programs and update your software. KPackageKit is fine for experienced Linux users, but Linux newcomers, which is what Ubuntu wants, will find it less than appealing. For more on Kubuntu, check out this excellent review.

More >

May 3, 2010
by sjvn01
0 comments

First, we kill all the patent lawyers

Actually, I don’t think we should kill all the patent lawyers. Some of my best friends are patent attorneys — no, really. But I’d happily stick a knife into the American patent system.

In the beginning, the U.S. patent system was meant to encourage inventors and innovation. Abraham Lincoln is reputed to have said, “The Patent System added the fuel of interest to the fire of genius.” That was then. This is now.

Unless the Supreme Court does the right thing and tosses out business practice and, by implication, software patents with the proper decision in the Bilski case, we’re stuck with a system designed to wreck anyone who actually tries to implement his own ideas.

You see, with many software patents there is no specific language, no hard code, but only descriptions of general processes that can be implemented in multiple ways. Now, you might think you could avoid patent trouble by looking up the appropriate patents and not using them. Good luck with that.

As Bradley M. Kuhn, then executive director of the Free Software Foundation, told me a few years back, it’s “difficult today to write any software program — be it free software or proprietary — from scratch that does not exercise the teachings of some existing software patent in the U.S.A.”

Back when Steve Ballmer, Microsoft‘s CEO, first started talking about how Linux patents might be violating Microsoft’s patents in 2004, Dan Ravicher, an attorney and executive director of the Public Patent Foundation, said, “There is no reason to believe that GNU/Linux has any greater risk of infringing patents than Windows, Unix-based or any other functionally similar operating system. Why? Because patents are infringed by specific structures that accomplish specific functionality.”

But let’s say you do search for existing patents. You think you’re clear of any possible problems, but then you get hit by a patent lawsuit anyway. And guess what: You’re in more trouble than ever. Why? Because now you could end up paying up to three times more in penalties because you might have been aware that what you were doing was in violation of a patent. With patent lawsuit damages already commonly running into the hundreds of millions of dollars, this kind of legal reversal is enough to kill all but the largest companies.

Is this a great country or what?

This is why Microsoft, despite being the loser in some whopping patent lawsuits, such as the $200 million-plus it owes i4i for violating its patents and the $1.5 billion it once owed Alcatel-Lucent, is happy to threaten other companies, especially those that use Linux or open-source software, such as Amazon and TomTom into licensing agreements.

You see, it isn’t a question of whether Microsoft’s patent claims are valid or not; that’s not important. What’s important is that if a company settles any potential patent worries beforehand with a few million, it avoids spending millions in legal costs and the remote chance of having to pay out hundreds of millions in damages.

God help any small company targeted by an industry giant with a patent complaint. The legal costs alone would ruin most of them.

What we really need is a complete overhaul of the U.S. patent system. If the Supreme Court doesn’t strike down business process patents, the federal government should take up the problem. As it is now, any software program can be attacked either by patent trolls — companies that do nothing but collect patents and then look for companies that might be implementing the ideas within them — or by big companies wanting to stomp out competition.

A version of this story first appeared in ComputerWorld.