Saturday, December 19, 2009

A Mac Mini DVI issue resolved

So, I had been waiting patiently for years to revisit MacOS X, having been left with a seriously bad impression due to (the now defunct) NetInfo. This past week I was ready to give it another go, and purchased one in the latest series of Mac Minis, intending to use my existing DVI wide-screen as a display.

The Mac Mini comes with a Mini-DVI to DVI converter, so I figured this wouldn't be a problem. But after connecting a single-link DVI-D cable from the convert to my display, the display failed to recognize the Mini's output. Using the same cable with another computer, and the display recognized the signal without issue. Moving back to the Mini: no dice. Argh!

After finding nothing useful in a cursory search of the 'Net and ready to throttle Apple, I ultimately tried resetting the PRAM per the manual, despite it not being listed as a way to resolve display issues. Fortunately, this did the trick.

I'm happy I still get to use my existing equipment (including my non-Apple keyboard and mouse). Time will tell if OS X proves to be a better experience this time around.

Thursday, August 30, 2007

Friday, August 20, 2004

Mounting ISO images as drives in Windows

After a brief search on Google, I found that Microsoft released an unsupported tool that can mount an ISO image as if it were a read-only drive.

Here's a link an article on the topic.

Here's another to the utilitiy.

... too bad there's no source code. I'd like to see a native port of vnconfig.

Monday, August 16, 2004

Indigo and Service Transport Neutrality

I've been meaning to do a rant on the IPC solution that RPC/HTTP provides, and how it is not adequate for certain applications. After reading Don Box's interview with .NET Developer's Journal, it seems we are finally starting to see a mainstream effort to remove our dependence on HTTP for web services. Who says "the web" is limited to port 80?

Don Box pin-points the idea that there is a need to move away from HTTP as a transport for services in favor of a transport-neutral stack. This is a feature both Microsoft's Indigo and the IETF's BEEP.

It's a great idea, and I suspect Microsoft will be adding all sorts of goodies to the Indigo framework SDK to make IPC under Indigo simple to implement.

.NET Developer's Journal Interviews Don Box

...on his role at Microsoft. Here's an excerpt:

.NETDJ: I'm perfectly happy using ASMX for Web services - why should I care about Indigo?

DB: ASMX is the closest to "the truth" of all our shipping technologies for distributed computing. If you are doing service-oriented programming, ASMX is the easiest to use and the most likely to keep you on the straight and narrow. You have to work harder to do the wrong thing. .NET remoting is more unbounded and happily allows you to do both right and wrong things with equal ease.

There is nothing wrong with ASMX. It has some limitations. One of the limitations is that the current implementation is bound to ASP.NET and pragmatically, that means it is really about working over HTTP. It can work over other transports, but there is no easy way to do that...some assembly is required. So, as people want to use other transports, ASMX won't be enough.

ASMX also has a very coarse-grained extensibility model. SOAP Extensions allow you to replace pieces of the plumbing, but only with a very coarse-grained tool. If I want to see the actual bits as they go in or come out, or the stack frame, there are integration points exposed for this. However, if I want to see the intermediate XML, I have a lot of work to do. For example, Web Services Enhancements (WSE), has to do some fairly obscure back-flips in order to fit into this model and to do XML-based processing. The ASMX/WSE user doesn't see this, but it's definitely more work for us than it should have been.

Friday, August 06, 2004

Preparing IIS 5.0 for .NET

For some reason, the first time I installed .NET 1.x on an IIS 5 server, it just worked. I don't know if this was because I had done the install via Windows Update, or if the likes of a phenomenon occurred. .NET applications ran fine.

This time around, I installed IIS and downloaded and installed the framework manually. ASP.NET applications failed to work at first go.

After tooling through the IIS configuration, I noticed that none of the .NET extensions were registered in the mapping subsystem of IIS (e.g. extensions such as .aspx, .asmx).

Apparently, to fix this problem, one must change to the .NET framework root directory and run aspnet_iisreg.exe -i, which installs the extensions, mappings, and whatever else IIS needs to run ASP.NET applications.

I haven't had a chance to use IIS 6.0 yet.

Sunday, August 01, 2004

10 Must Have Tools for .NET

It occurred to me recently that I needed something to compile .NET code snippets... "what was that tool called?" I knew an article had been written on some very cool and useful tools for .NET development, but for some reason I never bothered to post it. I'm cutting myself slack... I'm still learning to blog.

Without further ado, Ten Must-Have Tools Every Developer Should Download Now.

http://msdn.microsoft.com/msdnmag/issues/04/07/MustHaveTools/default.aspx

Tuesday, July 13, 2004

Installing pf on FreeBSD...

This is very crude, but:

1. download kernel source.

ftp://ftp.freebsd.org/pub/FreeBSD/releases/5.1-RELEASE/src/ssys*,install.sh

2. install kernel source.

./install.sh ssys

3. modify kernel, add these options.

# Packet Filter support
options PFIL_HOOKS
options RANDOM_IP_ID


4. rebuild kernel

config IMPETUS && cd ../compile/IMPETUS
make depend && make && make install


5. reboot

6. download the ports collection

7. install the ports collection

8. install the PF port

cd /usr/ports/security/pf
make && make install


9. install the PF startup script

cd /usr/local/etc/rc.d
mv pf.sh.sample pf.sh


10. add the following to /etc/rc.conf

pf_enable="Yes"
pf_logd="Yes"
pf_conf="/etc/pf.conf"


11. update /etc/pf.conf with your PF rules, e.g.

block in all
pass in quick on { lo0 }
pass out quick all keep state