Dear Netflix, University of Phoenix, and others who blatantly continue to use the blight of popup advertizing,
I wanted to let you know that I will personally continue to abstain from use of your services so long as you continue to use popup advertizing.
Thank you.
~____
It is quite common for me at work to be remoting into a computer working on it and realize I need to reboot it for one reason or another only the machine I'm remoting into is a client OS (Windows XP, Windows Vista, Windows 7) that has the reboot icon removed from the start menu in the remote desktop window.
If the machine is under my desk I can just switch over to it to reboot or hard reboot with the button on the front but sometimes the machine is in another room in another building in another city in another state or another country. Then I have no access to the physical machine.
On occasion if I'm using a Lab machine they will have some interface that I can use to reboot but not always. So what do you do?
It's actually quite simple, open up an elevated command window and run the following command:
Shutdown /r /t 0
The flag to restart is /r, then the flag /t plus a number sets the time delay before the command restarts the machine, a value of "0" reboots immediately.
For more information on this command run "Shutdown /?" from a command prompt.
One of the cool things about working at Microsoft is just how geeky people are. You can be in a meeting and there might be one or two people who follow sports (strangely it is possible they would be mostly women) and one of them would say something like how about that game last night and most of the people in the room would look at them confused thinking of World of Warcraft or some other sort of video game rather than a ball game.
Oh so much fun.
Have you ever come across the need to log the date and time a file was created only to discover dir doesn't do it?
I did. I spent a couple of days frustratingly trying to find a way to capture the information with little luck. Finally I found documentation in the call command that explained that "%~t1" expands %1 to date/time of file. This gave me enough information that I could come up with the following script:
@Echo Off
Call :FileTimeDateSize "MyDate" "MyTime" "MySize" "%WinDir%\notepad.exe"
@Echo:%MyDate%
@Echo:%MyTime%
@Echo:%MySize%
GoTo :EOF
:FileTimeDateSize
Call :AsignDateTimeSize "%~1" "%~2" "%~3" %~t4 "%~z4"
GoTo :EOF
:AssignDateTimeSize
Set %~1=%~4 & Set %~2=%~5 %~6 & Set %~3=%~7 bytes
GoTo :EOF
I was able to extract the date and time using "%~t3" but I wanted the date separated from the time so I wrote the second function to split out the date and the time by feeding "%~t3 to the function :AsignDateTime without quotes so that it will be split out into three tokens which are assigned independently to the desired variables to be used later.
If you run the script you can see that it is working because the date Notepad was built appears on one line and then the time on the next. I have also expanded this so that it reports the file size as well.
Now all I have to do to log the date and time so I can parse it later is @Echo:%MyDate%,%MyTime%,%MySize%>>LogFile.log
Woo! Hoo! I received the much anticipated Return to Microsoft email this morning, so I'm spending some time working on updating my Resume. Now how do I summarize the vast amount of random stuff I did in my last contract to only a few short sound bites?
In the previous article I introduced how to use batch to search and replace text in strings but what can you do with it?
Let's say you have a whole bunch of files that you want to work with but you are using some old program that frustratingly refuses to allow you to open files that have spaces in their names. Luckily this is a rare case now days but I recently came across such circumstances.
You could rename them by hand, but if you have lots of files it could get old fast. Believe me, renaming thousands of files by hand is a pain in the wrists.
The first possibility is that you could delete the spaces such as in the following:
DeleteSpaces.bat
@Echo Off
SetLocal EnableExtensions
@Echo Renaming all files in the folder to remove spaces.
For /f "delims=" %%a In ('dir /a-d /s /b *.*') Do Call :DeleteSpaces "%%~a"
GoTo :EOF
:DeleteSpaces
Set t=%~n1
Call Set t=%%t: =%%
Ren "%~1" "%~dp1%t%%~x1"
GoTo :EOF
But deleting spaces may not be what you want to do, after all you are deleting valuable information that spaces bring. What about replacing the spaces with underscores_ instead?:
UnderscoreSpaces.bat
@Echo Off
SetLocal EnableExtensions
Echo Renaming all files in the folder Replacing spaces with Underscores.
For /f "delims=" %%a In ('dir /a-d /s /b *.*') Do Call :UnderscoreSpaces "%%~a"
GoTo :EOF
:UnderscoreSpaces
Set t=%~n1
Call Set t=%%t: =_%%
Ren "%~1" "%~dp1%t%%~x1"
GoTo :EOF
What these both do is retrieve a list of tiles then use modifications of the find and replace function and replaces spaces either with nothing or an underscore then rename the files accordingly.
One thing I made certain to do is limit the renaming only to the filename and not the rest of the string. This is done by setting "t=%~n1" the n flag pulls only the filename. When the rename is done it is all put back together using "%~dp1%t%%~x1" the flags d and p are drive and path then %t% is the filename and it completes it using the flag x for extension.
note that SetLocal keeps the variables set in either script contained within the scripts so that they don't spill out into other scripts.
I wonder what this will do for the Amalgam universe?
http://www.marketwatch.com/story/disney-to-acquire-marvel-entertainment-2009-08-31-9050
I used to argue with my dad on the matter of precision. My dad is and engineer in education and personality as was his dad and his dad's dad and so on, so I inherited the mentality critical thinking and interest in mechanics, processes and technology of an that such a personality brings (Which sadly doesn't help with personal relationships:).
At one point I had decided I wanted to design my own game engine. I had decided I wanted to use the most precise decimal system I could find in 'cough' VB for storing the locations of points. In the process of designing the basics of the engine I tried to find pi to the most decimals that would fit in the desired unit. I had learned the 22/7 trick from my Drafting instructor but was completely unsatisfied with the accuracy of only 2 decimal points.
My dad asked why I thought I needed any precision beyond two decimal points; in the mechanical engineering world two or three decimals are usually considered accurate enough for almost any consumer products, as anything more precise makes little difference, and just the expansion and contraction of materials alone exceeds the difference such precision would bring.
But I was using a computer and despite what the 1950s and the movie industry hope to tell you, computers are inherently inaccurate at least when it comes to trying to calculate decimal and floating point numbers.
If you use floats for storing points in a world you can move around in, you would find that the precision of the locations of points in the center of the world is quite high, but the farther away you move from the center the less precision you have because the whole number portion of the value is getting filled with the larger numbers sacrificing the precision and storage space of the decimal portion. So at the edge of a large map you could end up with unexpected gaps in your world.
Well, okay if you were to only use integers the gaps could be reduced, but when you want to try to copy the nature of your surroundings you might find you want more precision. The thought might come up to use fixed point, sure it could work but you would have to be very careful with your rounding and your geometry. You also end up with geometry that can look a little blocky depending on the resolution of your values.
But I never got far enough along to be able to see any degradation. So now days, I've stopped caring. Well almost, when I go back and play with old geometry code I am still tempted to write everything as doubles rather than singles, but now I've learned to think processor capacity and the effects on speed too.
A while back I thought I would try to share some of my favorite batch script snippets from my Google notebook but the shared notebook was quickly flagged as potentially dangerous and I didn't feel like contesting the flag so I pulled the snippets back into my private notebook.
I still want to share a few of my most useful snippets, so here is my favorite which I have heavily adapted it to fit many needs
Replace.bat
@Echo Off
Set /p Source=Text you would like to search:
Set /p Search=Text you would like to find:
Set /p Replace=Text to replace "%Search%" with:
Set Dest=
Call :Replacer "%Source%" "%Search%" "%Replace%" "Dest"
@Echo.%Dest%
GoTo :EOF
:Replacer
Set t=%~1
Call Set t=%%t:%~2=%~3%%
Set %~4=%t%
GoTo :EOF
So what's going on here?
Well, first you provide a string to search, what you are looking for, then what you want to replace it with. The script then calls a function which replaces all instances of the search term with the desired replace term. Simple right?
The interesting part comes in the line which does the search and replace.
variable t is a temporary variable which gets set as the source string then it is replaced with the new string in a call set statement in this strange part "Call Set t=%%t:%~2=%~3%%" which could be written as "Call Set Destination=%%t:%Search%=%Replace%%%".
What it expands to is fascinating as it double expands. Say you want to replace "is" with "was" in the string "This is your search string". You would end up with "Call Set t=%%This is your search string:is=was%%" which needs to be expanded one more time as it is now an environment variable replacer which the call statement executes to replace the instance of "is" with "was" so you end up with "This was your search string".
The final confusing piece of code sets the result to an arbitrary variable name which you feed the function in this case "Dest", this allows you to reuse the code with any destination variable you would like.
This has been a fascinating howbeit confusing piece of batch script that I have found to be very useful in so many situations.
Now for a few notes about batch code hygiene in which this sample employs.
There are two graphics applications I use on occasion which automatically scan through photos and identify faces. They are Google's Picasa and Microsoft's Windows Live Photo Gallery. They both have their own unique uses and they are both getting better, but every time I move to a different computer or install a new operating system I find myself doing the same thing all over again.
On my little network hard drive I have tens of thousands of photos must of which are of people. Right now I can remember who many of them are, some I have forgotten their names already, but I want to make sure I keep track of who the people are. I've started a few times tagging people and adding comments to pictures, but then I go on to another computer only to discover the information was stored on the first machine only.
Windows Live Photo Gallery has built in tagging of people which is nice, the auto identification could be improved however. The manual tagging is almost as good as FaceBook. Actually it would be nice to be able to sync to FaceBook and other accounts to help identify who someone is, as it is now the only option is a Windows Live account.
Picasa doesn't seem to be able to locate people using just the installed application, but it can when the photo is put online and the identification online is actually pretty good, but I don't want to put all of my pictures online, I don't have enough space available for that on my account for that and I'm not wiling to shell out to get more right now.
What I wish would happen is they would write the tags directly into the photo's metadata so that it travels with the photo. And it would be nice if they used a method that would be interoperable between applications so that you could identify someone in one application and it could be associated as the same person in another application. Okay, that might be a near impossible idea but it would sure be nice.
Okay, I use Office quite a bit. At work Outlook, Excel and IE are my main tools, at home Word and Excel are among the top applications in my list. I write stories, keep track of my finances and hours and so on in these applications.
So when I found out that a couple of guys on the Office team decided to create a web page asking how Office could be improved; I jumped at the chance to try to get some of those pesky irritations I keep running into fixed in some future version.
So if you want to make Office better, just head on down to http://makeofficebetter.com to send in ideas suggestions and vote on things you want to see fixed.
I've been thinking (which is probably a dangerous thing) about what an operating system could look like if you could just start from scratch.
What if you could just ignore the existence of the current interface paradigms and created something new? What could it look like?
The current windowed paradigm goes all the way back to the old Xerox prototypes and everyone has been copying and improving on the idea ever since Apple, Microsoft, Unix, Linux, and the list goes on. Just about all mainline operating systems today package the jumbles of compiled code in boxes and call them programs.
I suppose unique is what you have with Singularity; I don't think I have seen an operating system that looks like it before, though it is pretty much all command line with a Matrix like feel. It's cool but most of the coolness is hidden behind the shell.
A cool design I have played with a little around work it the Surface PC which lends it's self immediately to intuitive interface design.
So, if you could design an OS and it's shell, what would it look like? Would it be 2D, 3D, Command line… what?
Watching what is going on in HTML development it might not be long until we have 3D web. Could it be the next OS interface design?
Would you have a game shell with lots of fancy graphics and cool effects or would you have a "keep it simple silly" style to boost processing to things that don't render on the machine to make a work server?
Hollywood has tried time and time again but the shells that are designed for TV and Movies are ridiculous at best designed for flashy effects rather than actual use. But they do present new ideas from time to time.
I'm not sure what I would do. The possibilities are limitless. I think 3D should be a native API, but not sure how it should be used.
Anyway, food for thought.
Alright, in three weeks I can start applying for jobs at Microsoft again which will give me five weeks that I will likely be interviewing about three times a week until the hundred days is up on the 8th of October. That means that I have three weeks of nothing to fill.
I did sign up for a two day ASP.Net class next week hopefully it is interesting.
Today After watching a little of Season 4 of Alf I couldn't stand being cooped up anymore and finally got out of the house and rode my bike around Robinswood park for about half an hour then down to the Chevron near Albertsons to fill up my tires (It's the only place I have found around here with free air).
After that institute. The lesson was continuing the previous subject and adding priorities and goals.
I'm having a hard time getting to sleep tonight (probably too much sugar in the strawberry milk) so I thought I would write out what is on my mind.
Classes vs. Structs.
In C# and VB.Net there are both Classes and Structures which can act fairly similarly, but the have differences that make one more appropriate in a situation and the other in another situation. For my little geometry program I have been debating back and forth on which to use for the shapes.
So far I have used Structs as it just seems natural that they should be. The other advantages is the natural garbage collection, and that they are initiated on the stack rather than the heap so they are lighter to fling around.
But there are disadvantages for example not being able to use inherited classes. This is what I'm debating over.
Classes are a little heavier and tend to be initiated on the heap so there is possibility that there could be some lag to their use. But they can Inherit other classes which means that you can create a base class with all of the display stuff packaged up into the class which other classes can inherit and use like it is their own. The other classes could take care of all of the other weird stuff but retain the same basic variables.
On the other hand both classes and structs can use interfaces. Interfaces are similar to inheritable classes except that they only define what a class or struct must contain to be interfaced as it's type. This means you have one interface with a whole bunch of heads defined but nothing else. It also means that in order for a class or struct to be defined as that type it must contain full definitions for those pieces which the interface defines. This can bloat the code a little.
So, Define them like I already have as structs and continue adding the rendering and graphical property code, or redefine as classes and split off the graphical property code to an inheritable class while retaining the rendering code in the interface?
I've only been thinking it through so far, the idea seems plausible classes might offer the advanced flexibility I'm looking for. But I'll have to see if such a hybrid works when I start playing with the code again.
Soon after I moved up here to Washington I was given MS Money as a gift to help me maintain my finances. It does budgeting, spending tracking, home inventory, bill tracking and so on. But for the most part I use it so see when Paychecks and debit transactions have been processed and to see the current "actual" balance of my accounts. Basically I'm using it as a check register.
I then copy the transactions into an Excel Spreadsheet so I can see the pretty line graph which helps me understand how much I actually have and what is available since the numbers mean little to me.
Well… My 2 year online update license ends next month and Microsoft has abandoned the entire line stating that as of June 30, 2009 it is no longer available for purchase and those who were able to get the updated version before that date would only be able to use the online resources until January 31, 2011 when they will be taken down. Though I did notice a new note dated August 5, 2009 that there will be a version of Money that will be released that won't require activation, but also won't have online features.
So I'm left trying to figure out what to do. Microsoft is working with the makers of Quicken to improve importing as an option. But I'm not so sure I want to use Quicken.
I have data that goes back to August 16, 2005 (I know, not all that far back but still) data which is no longer available from my financial institutions so what ever I use I want to keep that data.
What are my options?
I'm not so sure I want to trial a whole lot of these to find one or a combination that works right.
Then again, what if I just wrote up my own financial software and used some free service to monitor? Eh. Time will tell.
Conway's game of life is an amazingly addictive game http://www.ibiblio.org/lifepatterns/ or http://conwaylife.com/ the concept seems so simple but it creates such fascinating complexity. It's especially fun finding the patterns that actually increase indefinitely.
Watching the web browser statistics is fascinating. In one glance you can get an idea of what people are using to view the web and get an idea of what browsers to focus on for testing for the majority case. It is also interesting to see how a particular browser is being accepted by the community. The one I have been watching lately for the general case is StatCounter(http://gs.statcounter.com/#browser_version-US-daily-20080701-20090731) which shows a nice graph that I can watch.
It is fascinating to see that IE6 is still hanging in there. IE7 held high ground for quite awhile there but it looks like it is slowly being replaced by IE8 now. I'm a little surprised to see that FireFox is starting to loose ground. And according to the chart Chrome is out of the picture so I'm guessing the latest version of Chrome is actually part of the confusingly high Other category.
Now that was by version number, which does not show the brand loyalty. Looking at StatCounter by Browser brand (http://gs.statcounter.com/#browser-US-daily-20080701-20090726) IE remains in the lead but Firefox pretty much mirrors IE across the 50% line which is a fascinating trend.
The rest look like they are staying pretty much even except for Chrome which is slowly gaining ground, however it is hard to tell but it almost looks like Chrome is getting customers from FireFox.
Anyway fascinating statistics. It will be interesting to see what happens when IE9 comes out in a couple of years.
Every once in a while I will get frustrated with a website that I try to use and I send feedback in the form of a Bug report:
Descriptive Bug Title
------------------------------------
Short description of problem.
- Repro Steps:
- First thing to do
- Second thing to do
- And so on…
Expected:
The expected behavior.
Actual:
What actually happens when the repro steps are followed.
Recommendation:
Any recommendations to correct the issue if available.
I remember back to when the Internet was still new and domain names were being registered for the first time it was fascinating to watch as word after word was being gobbled up to assign new pages to. It looked like it wouldn't take all that long before every word in the dictionary and a huge portion of short phrases would be claimed. I figured that by the time I was an Adult they would all be claimed.
Not having internet connection at that time I used to write down every web address I wanted to visit on little sticky notes which I kept inside the flap of a calculator that I used to carry around, in the hopes that I might be able to spend a little time at a library, lab or somewhere else looking at these fascinating pages. Oh how things have changed.
So it is interesting then to see the results as an Adult now when indeed a good portion of the simple addresses have been claimed but there are still words and phrases left untouched and many of the pages once claimed have been left untended abandoned or taken down. Some domains have even been let to expire.
I wonder whet the next decade will bring?
