I have used a number of digital cameras, but they've never come even close to the quality of a decent 35mm camera. I recently have tested a Canon EOS Digital Rebel, and can tell you honestly that times are changing.
The EOS Digital Rebel takes fast pictures (as fast as 4-shot bursts), in quality levels that easily rival those of 35mm shots. It has all the manual settings you would expect form a good 35mm (aperture priority, shutter priority, bulb setting, etc.), and is the size and shape of a 'real' camera. The Lithium Ion battery didn't move from 'full-charge' through the whole weekend of testing, and a 512MB card looks like it will hold about 220 of the 2+ MB images (I didn't even get into shooting in 'raw' mode with 6MB images. Frankly, I don't see a need).
The camera I tested was paired with a Canon CP-100 dye-sub printer (out of production...replaced by the CP-200) which produces 4x6” prints in about a minute (each). You can connect the printer to a computer first, or print directly from the camera.
Offloading images from the camera is a breeze with the wizard that installs with the Canon driver...pick which images you want copied by selecting their thumbnails, and the wizard does the rest.
By the way, the printer supplies can be purchased on Amazon.com for about $14(US) for 36 prints. The quality is good enough that we're using this camera for fingerprint comparisons (under magnification). Honestly, you can't look at the resulting prints and tell that they aren't from the photo lab.
John Mandia threw down a cool challenge idea...how to make Google search multiple sites at one time for a specific term. I messed around with it for a while, and came up with this:
http://www.google.com/search?hl=en&lr=&ie=UTF-8&safe=off&as_qdr=all&q=bowen+john+%7Bsite%3Agobowen.com+OR+rainbowportal.net%7D
Basically, from the Google interface, use the following format:
bowen john {site:gobowen.com OR rainbowportal.net}
...Where 'bowen' and 'john' are the terms to search for, and 'gobowen.com' and 'rainbowportal.net' are the two sites I want searched. Neato! (though admittedly, this example is a bit conceited)
[edit]After posting this, I went back and read the comment from Clint S. on John's site - it looks like the curly brackets aren't needed for single-term searches, but for multiple terms, you have to have them[/edit]
[edit2]6/27/2004 - Wrong again...Clint S. had the right answer from the start. Still a cool trick, though.[/edit2]
In helping to convert a three-ring-bound procedure manual to something better, I’ve noticed that even though the Internet has been around for years, people struggle with parts of its concept
Background
The manual I’m talking about here is a very technical document. It details procedures for fingerprint examiners to follow in all aspects of their jobs. It’s existed in a three-ring binder for many, many years, and is critical to the job for reasons that aren’t important in this context.
Problem
The problem is that the document is ‘ordered’. Like most ‘normal’ documents, it has a table of contents, with chapters that group similar content together. Unfortunately, no one reads it that way. When you need it, you need a specific part of it, but no one reads the whole thing (or even a whole chapter) in the normal course of business. It’s a reference for those times when you can’t remember a very specific item or task.
Observation
There probably are some neat terms out there that describe this concept better, but I’m calling them ‘Ordered documents’ and ‘Unordered documents’ (and hereby claiming the rights to all profits that my ‘new’ terms generate<evil laugh/>). An ordered document is like a novel...you read chapter 1 in order to understand chapter 2, then on to chapter 3, etc. You don't (usually) skip around the book and read just certain parts of it unless your bookmark falls out.
An unordered document is more like our manuals (probably like most technical documents). We might go directly to the recipe for a Ninhydrin solution, then click from there over to safety protocols, then to standards and procedures, etc. We don't read through anything start to finish (well, most people don't anyway).
The web works like this…it’s littered with links in places that will take you abruptly to somewhere else. Many people intuitively understand this concept when they see it. But when writing documents, we struggle because the only way most of us know how to write documents is in 'order'. Putting a 'link' in the middle of a paragraph is still uncomfortable to many people. Worse yet is converting a document that has been ‘ordered’ for its whole life.
Solution
Yeah, right. There are a number of solutions out there, probably many I’ve never even heard of; for this case, I’m pushing DocBook (not Dita…sorry Jes!). I’ve spent a ton of time all over Bob Stayton’s site and have managed to squeeze out some good proofs of concept
I was recently dealing with a badly damaged installation of Windows XP Home...it wouldn't recognize its modem or CD ROM drive, wouldn't run System Restore (couldn't even get the window open to try it), and generally had suffered a lot of loss due to the 318 virus instances I had removed from it.
I decided to start clean. I ran the Files and Settings Transfer wizard, moving all the files I needed to a safe place...then dropped the partition, reformatted, and reinstalled XP.
With that done, I re-ran the F&S Transfer wizard to bring back the needed files...but found that that backup was only partial. I was missing some 490+ jpg files (family photos) from it!
Now, to the point (thanks for hanging on this long). After a bit of sweating, crying, and pounding of fists, I found RecoverMyFiles.com, which claims it will recover lost files even after a re-partition and reformat. Let me tell you that it does exactly that! I was able to recover all of those family photos in a couple of hours (the scan required after reformatting is a bit slow) without a problem.
I'm not usually big on endorsements, but I highly recommend RecoverMyFiles.com. It saved the day for me.
I was trying to map a network drive on PC A to PC B on a small network...with poor results. The Map Network Drive wizard kept complaining that it couldn't find the share I had set up on B...even though I'd used that share successfully dozens of times before with other machines.
I made sure that both machines were in the same workgroup, but that didn't do it. I finally disabled the Window XP Internet Connection Firewall on PC A and that did the trick. I thought that firewall was supposed to prevent traffic in only one direction (in), but apparently it prevents network activity in the out direction too.
I've never been so pleased to see Rainbow's CVS download broken! As strange as it may seem, it just reminds me how much work is getting done on the project, as developers from all over the place are squashing bugs at an incredible rate.
I suspect the CVS package will stabilize in the not-so-distant future, and Rainbow will be stronger and more stable than ever.
This weekend, my 4-year old and I were in the garage when he suddenly started crying out in pain. I went over to him, and he held out a spray can of “Deep Woods Off” insect repellant while screaming "I sprayed this in my eyes!"
We rushed him into the house, rinsed his eyes with water, then dried him off. After a few minutes, the tears stopped falling, he looked up and said calmly to my wife and I, "Remind me not to do that again."
With Rainbow installations, occasionally I'll end up with some (or all) objects in the database being owned by my local ASPNET account (computername\ASPNET). This results from building the database with a trustedconnection, and while I know this, I always forget and do it anyway.
The problem is that when I try to upload the existing database to a 'live' one somewhere, now that local ASPNET account owns everything and things go very sideways.
Anyway, I found this SQL script a while ago that (in two steps) changes the ownership of all the objects in a database to 'dbo'. I didn't write it, and it's been so long now, I don't even know where I got it...if it was yours, thank you very much. If not, here it is:
SELECT 'EXEC(''sp_changeobjectowner @objname = '''''+ ltrim(u.name) + '.' + ltrim(s.name) + '''''' + ', @newowner = dbo'')'
FROM sysobjects s, sysusers u
WHERE s.uid = u.uid
AND u.name <> 'dbo'
AND xtype in ('V', 'P', 'U')
AND u.name not like 'INFORMATION%'
order by s.name
Run this in Query Analyzer (with 'Results to Text' option set), then copy those results into a new QA window and run them.
On the off chance that someone else will encounter the same errors, I'm posting the problems I'm encountering in this upgrade.
After hearing about Google's new GMail product, where email is "categorized" instead of just jammed into folders, I was thinking it would be neat if you could do the same thing with files in the Windows file system.
Basically, you'd need a way to index the existing files, categorizing them according to the user's needs. For example, one Word document may be categorized in the 'Word documents', 'Code fragments', 'Memos', and 'Development Team' categories...instead of just getting filed away in 'My Documents' under, at best, a folder representing one of those categories.
The beauty of this would be in searching. Besides the usual 'search by name' junk, you could restrict a search to certain categories. (think...“I know I sent that code fragment in a memo to the whole development team“)
Using Google, I looked around for such a thing and found that Microsoft's Longhorn is packaged with exactly what I'm dreaming about: they call it WinFS and the article that dashed my dreams of fame and fortune is here.
I noticed this post from Ed Daniel a few days ago, but didn't get a chance to check it out until today. I really dig the format (short video interviews, quick and to the point).
A current Windows Forms project of mine requires the caching of a fair amount of data. Specifically, I want to store, in memory, on the client, a selection of XML data. When it's done, this should behave something like ASP.NET's session object...so I'll have access to user information and preferences without having to hit the data source every time to get it.
It's no problem to instantiate a class that retrieves and holds this information, but making that information available to other forms in the application (without again reloading an instance of the class and re-retrieving the data) is escaping me. I explored Microsoft's Caching Application Block, and while it appears to offer the kind of behavior I'm after, it's a bit more complicated than I need (and depends heavily on SQL Server...I want to avoid that).
Anyone have any suggestions out there?
It's a little like watching the neighbor kid grow up...a year ago, a Google search for 'Rainbow portal' would hit on rainbowportal.net, and not much more.
Now, you get pages and pages of hits, including this one and this one, which offer special Rainbow hosting packages. I can't wait to see those results next year, with a v1 release behind us and .Next getting rolling.
I agree with this guy...who agrees with this guy...
Although, as I look around, I have five different 'Teach yourself/Learn X in Y days/Hours' books on the shelves. *blush*
Every time I restart my XP Pro laptop, it downloads the same update (KB835732) and asks to install it.
Reference to an interesting article that tries to draw some conclusions about whether Diabetes is related to certain structural patterns of the ridges in the palms of the hands.
Browsing Google search results for 'fingerprint blog', I ran across this entry on an adoption site, where subject was told his prints 'failed' because
“Some people just have bad prints that can't be recorded at the level required to read them effectively.” While this is not impossible, it's very unlikely, and a few comments on that entry had also been given bad information. I felt obligated to post a comment myself:
“Fingerprints vary in quality from person to person, but it's likely that yours are fine (meaning, OK).
Generally speaking, 'blue collar' type workers (i.e. stone masons, builders) can have terrible fingerprint ridge structure...they're constantly wearing off the skin that forms the ridges. The same is also true for folks (again, generalizing) who shuffle a lot of paper (i.e. bank tellers, envelope-stuffers), because the paper has a similar effect.
In the *vast* majority of cases, though, bad recording of fingerprints is due to operator error...and the quality produced by ink/paper (done properly) still beats any computerized recording system in existence.”
I heard this exchange in the doctor's office last week:
Patient (walking back for his Vasectomy procedure): “Hey Doc. I hope you got a good night's sleep.”
Doctor: “Yep, I did. But I've been drinking all day.”
|