Your suggestion(s) needed
It's been awhile since I wrote for MSDN magazine. Now that I work for Microsoft, the whole financial situation changes. The short story is that I don't get paid for any articles or columns I might write for them. I'm not crying about this too much, as my personal life keeps me busy enough, and I'm inclined to spend my weekends hunkered over a keyboard. If I lean something cool, I'll use this blog to present it.
Still, MSDN has been good to me, and I'll continue to help them out. I recently got a request to write the back page for the December issue. The topic can be pretty much anything I like. However, the suggestion I got was “...branch into something controversial about software development that is really important to you.”
There's many things in software that are important to me, but I'd like to hear from you folks. Got any suggested topics that you'd like me to rant on?
Comments
Anonymous
July 27, 2004
A controversial topic : RAD versus the reliability of code : how .NET improves that (or not) ?
Typical problem : the software works fine on your machine, and it's impossible to anticipate that it won't work on someone else's machine. Example : default credentials (or lack thereof) when using Http web requests.Anonymous
July 27, 2004
A lot of grief is caused by not giving due thought while designing interfaces. If those interfaces are exposed to others, the pain caused is enormous. For e.g. take an example of something as obscure as SetErrorMode http://blogs.msdn.com/oldnewthing/archive/2004/07/27/198410.aspx
This is an age old problem. It would be interesting to see an article on such interfaces/APIs exposed by windows and how could they have been avoided.
-AmitAnonymous
July 27, 2004
I think it would be interesting to see some coverage of how integration programming will change with the release of Longhorn. Possibly expanding on issues discussed here:
http://msdn.microsoft.com/longhorn/understanding/books/migrationguide/?pull=/library/en-us/dnlongmig/html/intmiglongch04.aspAnonymous
July 27, 2004
First of all, congratulations and good luck in the future with the blog!
Perhaps discuss about products being broken because of Windows updates, such as SoftIce? maybe an interesting reason for such a problem in the past.. maybe an insight on how to handle this issue so it won't break in the future (from MS side or NuMega/Compuware side, or both?)..
Silly idea I know. But could be interesting..Anonymous
July 27, 2004
Well, SoftICE is a special case. When you have something that depends so much on internal kernel structure layouts, it's really to be expected that it breaks when a new version comes out.Anonymous
July 27, 2004
Possibily controversial topics:
Will .NET be the new MFC? Will some apps be developed in .NET while others are built with the Win32/64 API? Or will Microsoft deprecate the Old Ways until .NET becomes the only choice? (Should they?) What will cross-platform development be like in such a world?
How should ISVs support older platforms? I've had the best luck by developing on the older versions of Windows, and testing on the newer ones. Alas, Visual Studio .NET doesn't run on Windows 98. (Heck, I can't even get the .NET Framework 1.1 to install, which means I still haven't been able to try out the cool Logitech IO pen I got last Christmas!) Some of us are building innovative desktop apps for budget-minded consumers and cannot afford to overlook the pre-NT market. Are we supposed to stick with Visual Studio 6 and develop on Windows 98?Anonymous
July 27, 2004
I went into sofware development because I loved to write code.
These days, much of the code is generated, much of my passion for programming is dissipating away. I am not fired up by generatings wads of code and connecting them together. I'd much rather sew and knit.
I still have fun learning how things work, but at work, it's not fun anymore.Anonymous
July 28, 2004
Most of the articles in the MSDN magazine deal with articles about the next release of software.
It would be appreciated if you wrote about how to design for performance upfront when implementing solutions using .NET targeted towards a particular language such as C# or VB.Net.
There are some informative articles available, but most of them deal with enterprise server based solutions and they barely scratch the surface.
From what I have seen is that people are trying to use .Net slowly by first trying an non critical solution which is a simple windows forms based application and then studying the performance issues on GUI,DB etc. In many cases it is unacceptable.It is largely due to inexperiance and also due to unavailability of artilces that provide concrete information.
I would appreciate if you followed your earlier style, which is a mix of 70% on current issues such as performances measurement and tuning and rest on upcoming releases.
Thank YouAnonymous
July 28, 2004
Open Source OSes
You've made your living digging into the dark corners of windows and have been quite successful figuring out how stuff works. Talk about the (perceived) value of having access to the source (which you do now). How does that change your pov and is it as valuable as people think?
Hey, you asked for controversial!!!Anonymous
July 28, 2004
I echo "A Software Engineer trying to embrace .Net" exactly.Anonymous
July 28, 2004
How about an update to some of your earlier articles. One topic in particular occurs to me...you wrote about how SEH was implemented in Win32 - how about an article that goes into the details about how the .NET runtime implements its version of SEH on top of the underlying Win32 SEH. Related to this: how will the runtime achieve the same semantics on the newer platforms?
Another topic of interest would be another variation on an old subject. In Win32 there were various ways of intercepting calls to methods. With .NET you can use the profiling API to achieve something similar. How about an article on how to go about this?Anonymous
July 29, 2004
David: While I like the idea of giving an update on some of my earlier work, I only have a page to work with here. Plus, the back page is intended to be more commentary in nature.Anonymous
July 29, 2004
Should reverse engineering not only be legal, but be an essential part of developer skills, with tools like Soft-ICE, IDA and Reflector be used as widely as emacs?
Why should VisualStudio debugger not include a built-in decompiler that will step through 3rd party binaries in C#?
What does the future hold if developers have no right, will or skills to dissect the existing systems?Anonymous
July 29, 2004
The comment has been removedAnonymous
June 08, 2009
PingBack from http://hairgrowthproducts.info/story.php?id=5754Anonymous
June 19, 2009
PingBack from http://mydebtconsolidator.info/story.php?id=10715