Re: Factcheck.org is Useless

This is just my response to a recent post made by "Pudge" on his blog here: http://pudge.net/glob/2009/09/factcheckorg-is-useless.html

The particular example Pudge highlights is certainly indicative of the degradation of quality that factcheck.org has suffered, but it is by far not the most obvious or egregious example.

I found factcheck.org very useful several years ago, but over time began noticing discrepancies with other sources. At first I defaulted to siding with factcheck.org, but as I got more concerned with political matters, I started to do my own research, often going to the source; If an issue was over a bill in Congress, I would find the actual document and read it. If an issue was over a sentence from a longer speech, I would find the whole thing and listen to it.

It is my opinion that factcheck.org has gradually changed from a good source of objective, factual confirmations on the veracity of other claims in the media to yet another source of editorializing and subjective analysis.

This is not to say that factcheck.org is now as unreliable as the "mainstream media" - they still use more authoritative references and seem to still be far better on keeping claims in context - but I find myself having to parse and discard subjective opinions made by the author that are mixed in with otherwise objective information more and more often.

A disturbing trend I've noticed recently with factcheck.org is the apparent desire to sensationalize it's analyses and articles - headlines and leading paragraphs written not with the intent to inform but rather to grab the reader's attention and ride the coattails of controversy.

Personally, if I want this sort of research and reporting I would rather read the New York Times or the Wall Street Journal. If I want this sort of sensationalism I would rather watch Fox News or MSNBC.

BTW, yes I realize the irony of this comment - itself a subjective editorial. However, I do not claim this to be anything other than my opinion on the matter.

Notes

So... perhaps a blog would be a good place to put some notes

A few hours ago, I converted a CentOS 5.2 server to Debian Lenny. I've done it before, and it's not at all difficult - if you know what you're doing each step of the way. However, my brain is only so big and I can't remember everything so I made sure to find a few guides before I dove in.

http://www.void.gr/kargig/blog/2009/04/02/howto-remotely-install-debian-over-gentoo-without-physical-access/

http://www.howtoforge.com/how-to-remotely-install-debian-over-a-rh-based-distro-p4

The only problem is that neither of these guides are particularly up-to-date nor are they particularly straight-forward.

I'm making this post as a note to myself to write my own guide that can be more-or-less automated, but still offers the first-time debian converter the info they need to proceed with confidence.

Option parsing and documentation

After all this time programming, I am still dissatisfied with every command-line-option parsing and processing module/api/method available.

I've tried dozens in Perl, a handful in Python, a smattering in C, several in Java, and all fail to satisfy me in one or more ways.

The thing I find most dissatisfying is the API and usage of these option parsers. After that it's the ability (or lack thereof) to declare documentation and usage info in the same place where you define your option processing. After that, it's just that nothing seems to have all the features I want in one place.

I won't go into a tirade about everything I've tried and which system has what features and how awful or great each one is... I'll just work on explaining what I think would make up an ideal option parser and then try to lay out some ideas for API and/or implementation.

Example of usage of an option parser I'd like:



Presumably, this module would automagically generate --help output, usage docs, do basic error checking, and even output a man page or HTML or POD.