I always manage to get the fun ones.  Infragistics, Telerik, they all hate me.  Well, thats half true, I can force the Infa controls to do stuff most of the time -- haven't figured out how to make these Telerik.  So I'm working with winforms RadTreeView.  Seems simple enough.  Has a object-ed datasource  along with various things like "datamember" "displaymember" etc, all the things you'd expect.  Checkin out their site, you can use business objects, datasets, datatables, all sorts of junk -- what could possibly go wrong?

Well, we have one already in use, works just fine but we had to manually build the "nodes" for the control.  I think its a royal bastardization BUT it works.  So, thinking I'm mr. slick, I toss in a collection (a list) into the datasource, tell it that "object.property" is where the goods are and ...build, load, nada.  Hmm ok, soooo lets try a dataset.  I try using a helper class that serializes the thing into a nice neat dataset and ...nada again.  Hmm ok fine.  I'll build it by hand, using the nodes as cumbersome as it is -- nada.  Now, throughly confused, I think to myself "why not toss it into the one that works? see what happens" -- all the above methods work ...in the existing view. 

I've left a post on the Telerik site hoping for a reply along the lines of "hey, you're a dummy, all you have to do is this" and presto, it'll work ...but until then, I'm con-freaking-fused.

We were talking about this @ lunch a few days back, how people are now actually considering apple as a home computer because its "so much better" and the "cute commercials" with the business guy and the younger guy "I'm a mac" "and I'm a PC".  Our dicussion began to tail off into the secret world of "Mac's can do no wrong" and if you dare break any of the Mac commandments (this one in particular, "Thou shall not speakith against the mighty Apple!" not to be confused with "Thou shall not hack Apple products!")

Well guess what, people are actually complaining now ...and its not just a few of 'em.

http://scobleizer.com/2007/11/17/twitter-reactions-to-apples-os-update/#comments

http://discussions.apple.com/thread.jspa?threadID=1096520&tstart=0

http://www.tomsguide.com/us/2007/11/22/more_leopard_problems_plague_apple/

http://neowiki.neooffice.org/index.php/Leopard_Upgrade_Issues

http://www.macnn.com/news/48462

and a bunch others.

I swear, I don't go lookin for this stuff, but I'm very excited when I do.  Jeremey Allison is a huge supporter of open source code and in my opinion, does -NOT- like microsoft at all.  He's been cautiously outspoken them and their security practices which the obligatory response comes "who isn't?".  Well, interestingly (but not unexpected) enough, he's ran into a big surprise ...his code isn't secure and raised quite a stink.  I'm not dogging him for it, I'd be the first to admit my code isn't perfect (stop laughing!) but at least he's admitted it which begs the question -- did he tell on himself because there was no one else to blame?  Further, its written in C.  Do I think C is dead? Oh god no.  Is it dated?  Probably.  Do I think there's "better"?  Sure.  Is it practical?  mmmm, that's unclear with the answer of "probably not" (anyone got a c#/java version of samba sitting around?).

I find phishing scams amusing.  Always have, always will and I became worried for a while...I haven't seen anything different, new come across my email in a while.  Yes I have a life.  Moving on, I use to write tech articles regularlly for the office, and when I would get a "new" scam, I'd send it out and explain what all was going on and -why- this is stupid to respond.  Well, I found a new one!  This one is semi-new though -- consider it a repackaged "I'm a very wealthy foreign blah blah with a bajillion dollars I'd like to send you" ...but only slightly.  Here's the wording...(see the attached file as well, right here) 

We hereby confirm the receipt of part payment of US$8,500,000.00 dollars only, into our ESCROW vault / account with the HSBC Bank in London by order of the Foreign Debt Recovery Committee on Overdue Contract / Lotto Award / Inheritance payments in Africa / Overseas via the African Development Bank (ADB), as indicated in our database.

Yay!  How amusing.  Same kind of idea "we got money to give you now hand over your first born with a dna sample and your mothers madien name" sure to follow.  And to no surprise...

Kindly provide us with the below information to reconfirm your details in our database and for our prompt payment procedures.

Beneficiary's name / Age: ………………………………………………… Company's name / Rank in office: ………………………………………… Address:……………………………………………………………………

Tel: …………………………………………… Fax: ………………………

Receiving Bank: ……………………………………………………………

Address: ……………………………………………………………………

Bank Routing Number: …………………………………………………… Account number: ……………………………………………………………

Ding ding ding!!!  Congradulations!  You have successfully given someone just enough info to let your money go out the window.  Doesn't it feel good that 8.5mil is on its way?  This comment I find MOST amusing.

Once again, you are strictly warned against your further communication with your unscrupulous associates

What great advice not to share things with my not-so-careful "associates" because they might want some of my 8 million.  Well, I'm off to fill this out and next time you'll hear from me, I'll be on an island somewhere with a little umbrella in my drink!

I've been tasked with doing some investigational work regarding Com Wrappers for .net.  So far it seems fairly easy with a good number of articles around the web.  I've been able to find some good stuff right on the MSDN website along with few more sites explaining the operations/expectations.  The skinny of it is a client has a com object that does most of the talking between its external face and internal face -- now they're looking to create a stripped down version of the external face.  It should be straight forward, but we know how that goes.

*Update* Ok so I found some stuff in a certification book.  Talks about using a proxy (hmm, this sounds fimiliar - wcf what?) to create a runtime callable wrapper.  Using a tool called the Type Library Importer Tool to suck out all the meta data the COM file has and creates a .net proxy for it.  The command to do this is looks a little something like this.

tlbimp <somedll.dll> /out:<new.NETdll.dll>

You then add the new.NETdll to your project and you're done.

I just ran across this article this morning while I was doing the "getting latest" dance with TFS.  I haven't had a chance to really go though the article (although this project update is nailing my cpu for 100% during this entire blogging ...DataCompare is killing me) but the short of it appears there's a way to break the encryption based on "secret numbers" the algorithm knows about.  I might go as far as to tear into the 160 page doc and see what they're talking about but either way, "that ain't cool".

I needed a little humor, so I pulled up my favorite RFC documents.  Now before you pass out and die of bordom, hear me out ...and this was a stark reminder of how much of a geek I am.  Anyway there's two experimental RFC docs that exist, and yes they are real, and go as such...

A Standard for the Transmission of IP Datagrams on Avian Carriers

and... of course there's an update to an "old" 1990 protocol so we also have...

IP over Avian Carriers with Quality of Service 

For those of you who think I'm kidding ... check out this and this.  So whats the point?  Sometimes, RFC standards can be fun ...and god I'm such a nerd.  I think I'm going to suggest another update and suggest the transmission of datagrams using an affixed data-holding device for higher datagram capacity.

Ran into a really interesting gotcha some time ago and ran into it again.  It was rather irritating to go back and try to find the magical "how did I fix this?" so posting it was a good idea, and this is a great example of how something that makes perfect sense in code would produce a really strange, inconsistant "error" that you wouldn't expect.

So let's say you have a base class and there's a nullable datetime in there, "LastUpdatedDate" along with a nullable "CreatedDate" and a host of other junk that we don't care about right now.  The request is to retrieve the most recently updated item and at your disposal is a list of these items, we'll call it List<entry> entries.

entries.Sort(delegate(Enty entry1, Entry entry2)
{
     return entry2.LastUpdatedDate.Value.CompareTo(entry1.LastUpdatedDate.Value);
});

Well, bad news -- the LastUpdatedDate are equal.  How do you know when which entry comes out on top?  Did you say "that will never happen because of ticks, they will never be equal" ...not quite.  It's created on a database that you have no control over, IE its created on a trigger.  I ended up adding in another bit of logic to produce a more reliable output.  It isn't perfect.  Incase you are not fimiliar with CompareTo -- it returns a 1, a 0 or a -1.  1 is greater than, 0 is equal, -1 is less than, so we insert some logic into our anonymous method...

entries.Sort(delegate(Enty entry1, Entry entry2)
{
     int x = entry2.LastUpdatedDate.Value.CompareTo(entry1.LastUpdatedDate.Value);
     if (x == 0)
          return (entry2.CreatedDate.Value.CompareTo(entry1.CreatedDate.Value));
     else return x;
});

It's still possible that created date and last updated date could be equal, but the possibility is greatly reduced (in this case anyway).

For those of you who don't know, I have a degree in EET which more or less means I'm a much bigger nerd than I appear and every once in a while I see things on the web that really spark my interest on the EE side and this is very much one of them.

A quick overview of how phones work -- you are given a certain frequency range (3k I believe) which runs on top of a larger pipe (T1).  Within this pipe is a given frequency range and 24 channels, each channel has a guard band and 3k in each direction giving you a total of 8k of space (3k + 3k + 1k guard + 1k guard).  If you do the calculations, you'll find out how a T1 = 1.544mb.  Anyway, a nasty thing that can happen is cross talk.  This is when a given channel starts to flux outside of its given range and past the guard bands -- and thats very bad, you can begin to hear others conversations (who needs wire tapping eh?).  Anyway, when its running perfectly (or close enough), you'll never experience this, but it can happen.  Slight tangent, each channel is 64k which is also 1 voice channel and one channel of an ISDN line, sometimes called "fractial T1" depending on how is packaged (...or priced) but how you only get 56k over a phone line?  Hint - guardbands.

Anyway, ADSL works on regular phone lines at different frequencies, more specifically higher frequencies but there's a limit.  Well, this guy in Australia has figured it out.  I'm guessing he's using some kind of codec (encoder, kind of like what you use to create MP3s and videos) to jam more info into the same space with a little overhead.  Allow me to explain how this would work.  Say you have a text file with ...250k of info.  You would like to submit this file over some medium (flash drive, cables, whatever) but your limit is 35k, no more.  How would you do this?  Oh, there's a magical thing called "zip" that compresses that info into a much smaller space using a set of calculations/rules to reduce the size.  Yay that works, it cuts your file size down to under 35k and off your data goes.  When the recipient recieves this package however, it must know how you reduced (assuming it needs to, or more importantly, IF if even needs to uncompress) the payload.  If this is the only way to transmit, this is no problem otherwise another piece of verification must take place somewhere within the system, typically at the endpoint. As with anything, there's a catch -- the overhead.  Obviously, you shouldn't use a heavy heavy codec because that takes processing time and can slow down a system instead of speed it up.  Say for instance you are converting a movie of you took of your little ones into a file on your desktop.  Lets say native comes out to be 4gb (you're cool, you got an HD camera).  If you run it though a codec with some loss and convert it to ...700mb.  A good file reduction but now the codec to run it takes up 85% cpu to run.  Hmm, not a good trade off.  Another codec offers 1.2gb but only a 15% cpu usage when you run it.  Of course, you will have to consider your expected end point (is it a CD or DVD for instance).