We have a, shall we say, enthusiastically vigilant IT staff at work. Among other things, our computers are scanned 6 times a week for malware. I haven't worked out what scans what when, but there are at least 3 scans done as soon as the PC boots up and there are 3 that kick in right at noon. Leaving aside this kind of scanning frequency for a second, the problem we encounter is that Symantec's scanners max out CPU usage for about 90 minutes for each scan. It doesn't seem to matter how powerful the system is, the CPU usage is always 100%. So I'm left to conclude that the scanners are resource hogs and will take whatever's available.
If anyone knows of a task throttler that can force these gluttons to only devour 50% of my system resources, let me know. Right now, I just kill the processes as soon as they kick in. That's right, IT dudes: no scanning here. The alternative is for me to just pull the plug every time it kicks in. :)
Back to the frequency of scanning issue, am I alone in thinking this is a little excessive?
Monday, November 27, 2006
Monday, November 20, 2006
In the interests of full disclosure
This past Sunday's Meet The Press had interviews with two new Senators, Jim Webb of Virginia and Jon Tester of Montana. I'm not much of a partisan for any side-I'm a universal pessimist in that I think that no matter who's in charge I'm going to get hosed somehow. :)
One thing in the interview did stick with me though, the issue of transparency in government. Honestly, I don't think it matters what side of the political spectrum you fall on, doesn't it make sense that your representatives disclose their contacts? Their freebies? It seems only fair-every other level of government below more or less does it already.
One of the "perks" of The Day Job is the free coffee. It's pretty awful-tasting, so whether it's an actual benefit or not is up for debate. A cup of our coffee is like an airplane landing-any one you can walk away from is a good one.
Despite all this when we have a government visitor we frequently have to put a cup out to collect change. Why? Because they aren't allowed to accept so much as a free cup of coffee.
If the lower echelons of the Federal Government feel the need to account for even a 25 cent cup of coffee, is it really asking the higher ups all that much to be accountable themselves?
One thing in the interview did stick with me though, the issue of transparency in government. Honestly, I don't think it matters what side of the political spectrum you fall on, doesn't it make sense that your representatives disclose their contacts? Their freebies? It seems only fair-every other level of government below more or less does it already.
One of the "perks" of The Day Job is the free coffee. It's pretty awful-tasting, so whether it's an actual benefit or not is up for debate. A cup of our coffee is like an airplane landing-any one you can walk away from is a good one.
Despite all this when we have a government visitor we frequently have to put a cup out to collect change. Why? Because they aren't allowed to accept so much as a free cup of coffee.
If the lower echelons of the Federal Government feel the need to account for even a 25 cent cup of coffee, is it really asking the higher ups all that much to be accountable themselves?
Monday, November 13, 2006
RFI: Stamp Press Inspection
I have a request today for any U.S.-based (Missouri preferred) NDT service providers that can inspect stamp presses:
"I need to have nondestructive testing done on eight stamping presses toIf you're interested, leave a reply to this post. Or feel free to contact me offline (contact info at the bottom of this post).
test for fractures in the frame and drive train components. The presses
are 6, C-110 Bliss presses and 2, 5-40 toggle presses."
Java GPL'd
In case you didn't know, Sun's releasing Java under the GPL today.
I know a lot of people are rejoicing about this, but frankly I wonder if this really is a good thing for Java or not. Not from a technical point of view, mind you, but in the uncertainty it might create.
1. Forking
Let's face it, Java will fork. Over technology, philosophy, language, etc. It'll happen. I don't think that this will be a major concern for veteran Java developers-after all, they're used to dealing with multiple systems (Sun, IBM, etc.) by now. It might give pause to management considering what development tools to use in their Grand New Project, though. Even if you as a developer know better, the PHBs of the world sometimes do not.
2. Confusing Licensing
Although my take on the licensing is that you'll still be able to develop proprietary (aka "closed source") applications, I find the fine details and the discussion/debate confusing. I'll convince myself that closed source apps are still a-ok, then I'll read another opinion that half convinces me the other way.
The new licensing is supposed to work much like the GNU Classpath License, which (if I'm right) allows for closed source development. I'm not a lawyer but it looks as though I might have to start playing one considering Sun's take on the subject. Worse yet, this response suggests not all the JDK components will be under the Classpath Exception!
Look, I know that according to Sun's FAQ on the subject that they'll continue to make and distribute a commercial JDK, so if you're truly terrified of the GPL you can stick with what you know. Actually, come to think of it this might drive people to Sun's VM over anyone else's...I wonder if Java will ultimately fork into Commercial and Free versions?
3. Conclusion
Despite everything I said, bottom line I think anything that encourages more people to use Java is a very good thing. I plan on steering clear of the forking/licensing thing entirely until the dust settles and only using / targeting Sun commercial Java.
I also plan on taking a look at C#.
Update November 13 2006 19:08 CST: David Illsley has an interesting take on the JDK distribution question. Hopefully some of these questions will be addressed in the (very) near future.
I know a lot of people are rejoicing about this, but frankly I wonder if this really is a good thing for Java or not. Not from a technical point of view, mind you, but in the uncertainty it might create.
1. Forking
Let's face it, Java will fork. Over technology, philosophy, language, etc. It'll happen. I don't think that this will be a major concern for veteran Java developers-after all, they're used to dealing with multiple systems (Sun, IBM, etc.) by now. It might give pause to management considering what development tools to use in their Grand New Project, though. Even if you as a developer know better, the PHBs of the world sometimes do not.
2. Confusing Licensing
Although my take on the licensing is that you'll still be able to develop proprietary (aka "closed source") applications, I find the fine details and the discussion/debate confusing. I'll convince myself that closed source apps are still a-ok, then I'll read another opinion that half convinces me the other way.
The new licensing is supposed to work much like the GNU Classpath License, which (if I'm right) allows for closed source development. I'm not a lawyer but it looks as though I might have to start playing one considering Sun's take on the subject. Worse yet, this response suggests not all the JDK components will be under the Classpath Exception!
Look, I know that according to Sun's FAQ on the subject that they'll continue to make and distribute a commercial JDK, so if you're truly terrified of the GPL you can stick with what you know. Actually, come to think of it this might drive people to Sun's VM over anyone else's...I wonder if Java will ultimately fork into Commercial and Free versions?
3. Conclusion
Despite everything I said, bottom line I think anything that encourages more people to use Java is a very good thing. I plan on steering clear of the forking/licensing thing entirely until the dust settles and only using / targeting Sun commercial Java.
I also plan on taking a look at C#.
Update November 13 2006 19:08 CST: David Illsley has an interesting take on the JDK distribution question. Hopefully some of these questions will be addressed in the (very) near future.
Sunday, November 12, 2006
Vista And Developers
I've been thinking about what the imminent Vista drop will mean for developers as of late.
While some of the hand-wringing for the User Account Control feature is probably a bit overdone, I do think there is some cause for concern. For example, the Java developers out there might want to check out SE 6's Vista Notes to see what's new. Developers in general should check out Microsoft's Application Compatibility Cookbook for a good rundown on the potential problems and workarounds.
Coming from a Linux background these new security controls don't seem all that onerous to me, so I think the main issue for developers will be how long it will take end users to get used to UAC and the other security features in Vista. Some commentators are worried that Joe Average won't know how to run as an administrator and won't bother to run/buy their applications if it's a required step. What's forgotten though is that there will be a ton of applications in this same situation: at some point if Joe wants to run anything, he'll have to bite the bullet and figure it out or get someone to do it for him. In other words, Joe'll come across some app he simply must have and learning to install that will be the door opening for the rest of us.
Remember, there was a lot of doom and gloom talk when XP's user accounts were first announced too.
I think that Vista is a great opportunity for REALbasic developers (and web developers for that matter), since RB apps are normally self-contained...no DLLs or runtimes to worry about under UAC. If you can tailor your application so that it doesn't require global R/W access, or anything else that runs contrary to the new Vista security model, you might have a leg up on your Java/C++/other language of choice competition.
Speaking of Java, while it's heartening to hear that over 60% of all new PCs ship with a JVM, it's going to be tough to beat 100%: every new Vista machine will come with .NET installed. I've never really taken a look at C#, because Java did everything I need it to do.
(That and the fact that developing commercial apps under Java is essentially free, but with C# you have to buy the obfuscator, new UI controls, etc. For a lot of SMBs that aren't development houses, free is important.)
Now I find myself looking more towards C#. From my experience with delivering Java desktop apps I know that just because "a" runtime is installed it doesn't make it "the" runtime you need, but it's a step in the right direction. Now with Vista's release, coupled with Microsoft's partnership with Novell and its ramificiations for Mono, it may finally be time for me to take a look at C#.
Anybody know of any free C# obfuscators?
While some of the hand-wringing for the User Account Control feature is probably a bit overdone, I do think there is some cause for concern. For example, the Java developers out there might want to check out SE 6's Vista Notes to see what's new. Developers in general should check out Microsoft's Application Compatibility Cookbook for a good rundown on the potential problems and workarounds.
Coming from a Linux background these new security controls don't seem all that onerous to me, so I think the main issue for developers will be how long it will take end users to get used to UAC and the other security features in Vista. Some commentators are worried that Joe Average won't know how to run as an administrator and won't bother to run/buy their applications if it's a required step. What's forgotten though is that there will be a ton of applications in this same situation: at some point if Joe wants to run anything, he'll have to bite the bullet and figure it out or get someone to do it for him. In other words, Joe'll come across some app he simply must have and learning to install that will be the door opening for the rest of us.
Remember, there was a lot of doom and gloom talk when XP's user accounts were first announced too.
I think that Vista is a great opportunity for REALbasic developers (and web developers for that matter), since RB apps are normally self-contained...no DLLs or runtimes to worry about under UAC. If you can tailor your application so that it doesn't require global R/W access, or anything else that runs contrary to the new Vista security model, you might have a leg up on your Java/C++/other language of choice competition.
Speaking of Java, while it's heartening to hear that over 60% of all new PCs ship with a JVM, it's going to be tough to beat 100%: every new Vista machine will come with .NET installed. I've never really taken a look at C#, because Java did everything I need it to do.
(That and the fact that developing commercial apps under Java is essentially free, but with C# you have to buy the obfuscator, new UI controls, etc. For a lot of SMBs that aren't development houses, free is important.)
Now I find myself looking more towards C#. From my experience with delivering Java desktop apps I know that just because "a" runtime is installed it doesn't make it "the" runtime you need, but it's a step in the right direction. Now with Vista's release, coupled with Microsoft's partnership with Novell and its ramificiations for Mono, it may finally be time for me to take a look at C#.
Anybody know of any free C# obfuscators?
Wednesday, November 08, 2006
What The Deuce Is Nondestructive Evaluation?
I realized today that I've been throwing around terms like "nondestructive evaluation" or "NDE" without actually bothering with definitions. Time to correct that particular oversight.
Nondestructive evaluation (NDE), also known as nondestructive testing (NDT), nondestructive inspection (NDI), etc., is just a catch-all for any method of physically examining something without damaging it (taking samples, cutting into it, etc.). In materials sciences, we're mainly looking to gauge the physical condition of a structure. Things like whether it's corroded or rusted, whether it has cracks or holes, that sort of thing.
As you may have already guessed, destructive evaluation is a process of physically examining something that also causes damage to the object in question. In general, destructive analysis is usually much more accurate than NDE, but of course that's at the cost of damaging the object. Damage is a subjective thing though and not everyone agrees on what's destructive and what isn't-take hydrostatic testing for example. To test whether or not a SCUBA tank is safe, it's pressurized to say twice its maximum pressure. Conventional wisdom is that a hydrostat is not a destructive test. My take is that it is a destructive test because by over-pressurizing the bottle you can create damage, e.g. if a tiny hairline crack forms.
My favourite analogy for NDE has always been the doctor's office. If your doctor thinks you might have broken your leg and orders an X-ray, that's nondestructive evaluation. If your doctor opts for exploratory surgery, that's destructive evaluation (cutting into your leg).
You can get more info on NDE from Iowa State's NDT Resource Center. A good public forum for NDT-ers is NDT.net. I'm a little hesitant to add Wikipedia's entry on NDT, but there it is. The page itself isn't very useful at the moment as it seems to have been hijacked for lack of a better word, but the links are useful anyway.
So there you have it. Hopefully you won't see "NDE" and think "near death experience" any more. :)
Nondestructive evaluation (NDE), also known as nondestructive testing (NDT), nondestructive inspection (NDI), etc., is just a catch-all for any method of physically examining something without damaging it (taking samples, cutting into it, etc.). In materials sciences, we're mainly looking to gauge the physical condition of a structure. Things like whether it's corroded or rusted, whether it has cracks or holes, that sort of thing.
As you may have already guessed, destructive evaluation is a process of physically examining something that also causes damage to the object in question. In general, destructive analysis is usually much more accurate than NDE, but of course that's at the cost of damaging the object. Damage is a subjective thing though and not everyone agrees on what's destructive and what isn't-take hydrostatic testing for example. To test whether or not a SCUBA tank is safe, it's pressurized to say twice its maximum pressure. Conventional wisdom is that a hydrostat is not a destructive test. My take is that it is a destructive test because by over-pressurizing the bottle you can create damage, e.g. if a tiny hairline crack forms.
My favourite analogy for NDE has always been the doctor's office. If your doctor thinks you might have broken your leg and orders an X-ray, that's nondestructive evaluation. If your doctor opts for exploratory surgery, that's destructive evaluation (cutting into your leg).
You can get more info on NDE from Iowa State's NDT Resource Center. A good public forum for NDT-ers is NDT.net. I'm a little hesitant to add Wikipedia's entry on NDT, but there it is. The page itself isn't very useful at the moment as it seems to have been hijacked for lack of a better word, but the links are useful anyway.
So there you have it. Hopefully you won't see "NDE" and think "near death experience" any more. :)
Tuesday, November 07, 2006
Ordering Scientific / Technical Articles
By the way, if you've just found a technical article of interest and need to order a copy, I highly recommend CISTI (Canada Institute For Scientific And Technical Information). I've used them a number of times over the years and they're just about the most inexpensive source for technical publications you'll ever find.
Scientific And Technical Information Network
The entity formerly known as NTIAC has one of the most extensive nondestructive evaluation libraries in the world, at last count more than 65,000 entries. Generally speaking, if you're in the market for NDE information, you contact us and we perform the search for you, as the library isn't generally available online.
Or at least it wasn't, until I noticed last week that STINET (Scientific and Technical Information Network) has a free public interface that covers most of the NTIAC holdings. Some of the restricted documents aren't there, along with new entries from the last 2 years or so as far as I can tell, but other than that the lion's share of the NTIAC database is up and ready for your perusal.
STINET covers more than NTIAC of course, but I'll leave that up to you for experimentation. If you've ever been frustrated trying to do a technical article search on conventional search engines, it's definitely worth a look.
Or at least it wasn't, until I noticed last week that STINET (Scientific and Technical Information Network) has a free public interface that covers most of the NTIAC holdings. Some of the restricted documents aren't there, along with new entries from the last 2 years or so as far as I can tell, but other than that the lion's share of the NTIAC database is up and ready for your perusal.
STINET covers more than NTIAC of course, but I'll leave that up to you for experimentation. If you've ever been frustrated trying to do a technical article search on conventional search engines, it's definitely worth a look.
Monday, November 06, 2006
Mini-Review: REALbasic 2006r4 Linux
Whoops! In my haste to do a REALbasic Mini-Review I forgot to say much of anything about the Linux experience!The REALbasic IDE GUI builder is actually fairly sluggish in Linux in my experience. Granted, this may be because I run less than optimal Linux workstations at the office, but I should point out that NetBeans' Matisse is infinitely more spritely under the same working conditions. Still, the RB IDE loads much faster than NetBeans, but NetBeans comes out ahead when it comes to running apps from the IDE.
Other than the general sluggishness of the IDE the experience in Linux is about the same as it is for Windows. In my experience with both platforms, there's one aspect of XP work that's mercifully absent in RB Linux, and that's debugger-firewall collisions. The RB debugger stub uses a network interface to communicate with the IDE, and as a result I've yet to find a firewall under XP that didn't complain. The problem stems from the fact that even though you can generally grant permission to use certain ports and IP addresses, your firewall will still notice that the application in question (the one you're trying to debug) has been "altered." If anyone out there has a better approach let me know, but right now I tend to just disable the firewall while in REALbasic. Sub-optimal. Thankfully, Linux development in REALbasic doesn't seem to have the same hangup.
The applications built for Linux in REALbasic perform quite well, on par with a "conventional" application. There is a bit of a startup lag but it's not very bad, certainly less than that observed for a standard Java app. I've used RB Linux apps in embedded spaces for a while now and even on smallish devices the lag isn't a show-stopper. The Linux apps tend to be around the same size as their XP brethren, 2.5MB or so. Not a major problem, but they do stand out in the dynamic linking world of Linux.
If you're looking for comparisons with other RAD tools under Linux, I'm probably biased but I find the REALbasic GUI builder to be far superior to most of the other tools I've used. That's a personal choice for each of us to make though, but suffice it to say that if you're not a fan of the Matisse or .NET schools of RAD REALbasic isn't going to impress you either. If you are a fan or you're open to trying something new, it's worth a look.
All in all I stand by my earlier recommendation for REALbasic. Under Linux RB faces some pretty stiff competition (not to mention some pretty entrenched opinions about commercial closed-source applications) but I think it holds its own and will only get better with time. Since it's free under Linux, you might as well take a look even if you're inclined to disagree!
Sunday, November 05, 2006
Amazon 2.0?
I was all set to write a snarky post about Amazon needing a "show only Amazon.com" items filter, when I noticed it was already there under the "Narrow By Seller" heading. So instead I'll just say why I was looking. :)
I'll be the first to admit that I'm frequently hit by the dark lure of free shipping. Yes, I know that it would cost me less in most cases to just "bite the bullet" and give UPS/FedEx their due. There's just something about "paying for nothing," though. If I pay $5 for shipping, I have nothing to show for it. Except for the book at my doorstep, of course.
On the other hand, if I add $7 to my cart, I still get the book at my doorstep but now he's brought a friend! It feels as though I'm getting more for my money. With that in mind, problem number one with other sellers on Amazon is they don't count towards the free shipping I'm forever striving for.
Problem number one is compounded by the myriad shipping and return policies you have to navigate when you look at Amazon sellers. It's just like eBay in this, and the main reason I frequent Amazon is to avoid the eBay experience entirely.
Speaking of eBay, this leads us to problem number two with Amazon sellers: trust. I know what to expect and what I'm in for when I buy from Amazon, not so with third parties. I know 99% of them are above board, quality retailers. But again we're back to the eBay experience I'm trying to skip. I've bought many items from third party retailers through Amazon and I've never been burned or less than satisfied, but it was still only as a last resort that I went that route. There's always that twinge of uncertainty in the back of your head, isn't there?
So. Web entrepreneurs, here's your opportunity. Develop a marketfront for third party retailers that'll allow for free shipping for customers, even if they add multiple items from different vendors. Thoroughly vet any prospective vendors, and guarantee the customer's satisfaction.
Profitable? I have no idea. I assume you'd have to charge a premium to cover both problems, so you'd have to compete on quality of service and not on price. Still, there might be a certain percentage of the population that'd take you up on it, I know I'd take a look.
I'll be the first to admit that I'm frequently hit by the dark lure of free shipping. Yes, I know that it would cost me less in most cases to just "bite the bullet" and give UPS/FedEx their due. There's just something about "paying for nothing," though. If I pay $5 for shipping, I have nothing to show for it. Except for the book at my doorstep, of course.
On the other hand, if I add $7 to my cart, I still get the book at my doorstep but now he's brought a friend! It feels as though I'm getting more for my money. With that in mind, problem number one with other sellers on Amazon is they don't count towards the free shipping I'm forever striving for.
Problem number one is compounded by the myriad shipping and return policies you have to navigate when you look at Amazon sellers. It's just like eBay in this, and the main reason I frequent Amazon is to avoid the eBay experience entirely.
Speaking of eBay, this leads us to problem number two with Amazon sellers: trust. I know what to expect and what I'm in for when I buy from Amazon, not so with third parties. I know 99% of them are above board, quality retailers. But again we're back to the eBay experience I'm trying to skip. I've bought many items from third party retailers through Amazon and I've never been burned or less than satisfied, but it was still only as a last resort that I went that route. There's always that twinge of uncertainty in the back of your head, isn't there?
So. Web entrepreneurs, here's your opportunity. Develop a marketfront for third party retailers that'll allow for free shipping for customers, even if they add multiple items from different vendors. Thoroughly vet any prospective vendors, and guarantee the customer's satisfaction.
Profitable? I have no idea. I assume you'd have to charge a premium to cover both problems, so you'd have to compete on quality of service and not on price. Still, there might be a certain percentage of the population that'd take you up on it, I know I'd take a look.
Thursday, November 02, 2006
Mini-Review: Voq Professional Smartphone
Summary: probably the best new smartphone you'll get for less than $150.Today's little review is on a much-maligned smartphone, the Sierra Wireless Voq. I say much-maligned, because while "official" reviews have generally been positive, let's just say that the user comments in said reviews tend to be less forgiving. I've been using one for months now and thought it was time to post my experiences.
I'd have to say that I actually like this phone: it feels solid, it does what I want it to, it's a great intro to the whole smartphone concept, and it's cheaper than most other unlocked phones you'll come across with half as much functionality. What more could you want?
Other than WiFi, Bluetooth, etc. Still, it's a good office phone for this very reason. Without all that connectivity, camera, and other gewgaws, the battery life is better than some other smartphones too. I won't lie to you though: you will get looks using this phone. I convinced my boss to pick one up a while back-his experience has been that gawkers peg it as a phone from the 90's. Until you open the keyboard and start typing, that is.
The Voq was cancelled in 2005 due to lack of consumer interest. Back then it was a pricey little number ($600 U.S.), and if you do any searching on the Voq most reviews you'll come across will say something along the lines of "Nice, but not for that price." So any Voqs you find today are leftover stock. The good news is that they're very inexpensive for a smartphone-Tiger Direct frequently runs rebate specials that put it under $100 U.S. and Amazon is currently selling it for $119 U.S. Consider that new unlocked RAZRs are still in the $199 category, and you're in the right frame of mind to look at this phone.
The highlights:
- It is a little on the chunky side for a phone, but it doesn't bother me to wear it on my belt because it isn't very wide. In fact I notice the RAZR more than the Voq in the same situation.
- Accessories can be hard to find, but eBay searches are usually productive.
- As far as connectivity goes, forget Bluetooth and WiFi, it's not going to happen. It has a SD/MMC slot but it won't support SDIO cards so I hope you're ok with IrDA (infrared port) and USB tethering.
- The fold-out keyboard is a little cramped compared to other smartphones, but other smartphones tend to be wider as a consequence. Once you're familiar with it though the Voq's keyboard is very easy to handle.
- Sound quality is excellent and cellphone reception is terrific, much better than my RAZR.
- Battery life is fairly decent for a smartphone. I tend to need to recharge about once a week or so with fairly minor call times but with push email activated.
- Internet (GPRS only) connectivity is quite good, something of a headache to set up with your provider but not much worse than setting up any other unlocked cell phone.
- Push email works great and works with a lot of different email clients and applications. I can't believe it works with Groupwise-nothing ever works with Groupwise!
Closeup of the fold-out keyboard:
Size comparison with a standard RAZR phone:
NASA Seeks NDE Scientist
I'm a little late in posting this, but in case you haven't seen it yet, NASA's Langley Research Center is looking for a NDE Discipline Expert. Translated, they're looking for an engineer or scientist with an expertise in nondestructive testing and evaluation.
NASA's Engineering Safety Center (NESC) is a first-rate facility with a lot of very smart and talented people. Believe me, you could do worse!
NASA's Engineering Safety Center (NESC) is a first-rate facility with a lot of very smart and talented people. Believe me, you could do worse!
Subscribe to:
Posts (Atom)