Tags: cryptography

Horrible tags on Stack Overflow

Every once in a while I stumble about a tag on Stack Overflow which

  • has no useful meaning; or
  • is a marker for questions which are generally off-topic or not constructive.

Sometimes I try to open a post on Meta Stack Overflow about these to collect some help to clean them up. But as each user has only a limited amount of close-votes (and you need five votes from different users (or one from a moderator) to close a question), one can do that only every so often, these questions would take votes away from each other.

So here I’ll try to have a list of these tags so I don’t forget.

  • [future]: A lot of these are questions about future events. Prophesies are not on-topic on SO.. The on-topic meaning would be the concept used in concurrent programming.
  • [protected]: About quite some different things, like the protected keyword in some programming languages, protected memory, protected Twitter accounts, password-protected files, protecting video downloads, … . I’m not sure how to sensibly retag this. (The tag also has no tag wiki.)
  • [alternative]: This looks like another tag which contains many non-constructive questions.

  • [documents]

  • [word] vs. [msword]

I’ll add more later as I stumble about them.

Die “Ing-DiBa” Bank hat gerade eine Kampagne laufen, bei der an 1000 Vereine für gemeinnützige Projekte je 1000 Euro verteilt werden.

Einer der Kandidaten ist der Deutsche Esperanto-Bund, welcher damit Reisekosten von ausländischen Teilnehmern zu Esperanto-Treffen in Deutschland bezuschussen will.

Um dafür ausgewählt zu werden, müssen viele Leute dafür stimmen. Dazu muss man auf der Webseite der Ing-DiBa seine E-Mail-Adresse eingeben (und noch ein Captcha lösen, um zu zeigen, dass man nicht blind ist), und dann in dem in der erhaltenen E-Mail enthaltenen Link seine Stimme bestätigen.

Man darf pro E-Mail-Adresse bis zu drei mal abstimmen, für den gleichen oder auch verschiedene Vereine. Der DEB ist zur Zeit auf Position 1882. Die Aktion läuft noch bis zum 15. November, und wir hoffen, dass wir dann mindestens auf Position 1000 sind.

Tags: git dvcs english

The standard answer to embedding source code listings in a LaTeX document is the listings package. It works in pure LaTeX, by parsing the source code and highlighting some parts of them.

An alternative I just found on Stack Overflow is the minted package, created by Konrad Rudolph. It uses the Pygments library to do the actual parsing and highlighting.

I did not yet try the package, but I will and then update this post for my experiences.

It shows that each of my seven browsers here has a unique fingerprint.

In most of them, even the Accept header is already unique (there is Esperanto as first language, followed by German and English).

By Jeff Moser, this is a guide in stick-figure-comic form, which explains how AES came to be, how the AES cryptography works, and also some of the math behind it.

It also mentions “Confusion” and “Diffusion”, the important principles in Cryptographic algorithms.

There is a corresponding implementation at GitHub.

Tags: english names

An article about how to model hierarchical data in a relational database (specifically MySQL, but likely also applying to other relational DBMS).

Originally, this article was on the MySQL developement site, and linked heavily as http://dev.mysql.com/tech-resources/articles/hierarchical-data.html. But now Oracle (which owns MySQL after buying Sun, which itself bought MySQL some years ago) seems to have reorganized its site, with the effect that this link goes only to an Oracle search page.

This was remarked by some anonymous Stack Overflow user, who suggested the following edit to an answer linking to this page:

— UPDATE: The link is dead. Thanks Oracle. —

I saw this edit suggestion by browsing through the list of suggested edits, and thought this article should not have been completely gone away.

Googleing the URL did not really result in finding the current location of the article (only in some pages linking to the original one). A good idea in such cases is the Wayback Machine of the Internet Archive. It stores previous versions of lots of pages - this one, too (last accessed in January 2010).

If I only wanted to read the old article, this would be enough - but no, I in fact want to correct the link, and it is not a good idea to link to the Wayback machine. But having the article itself, it was easy to copy a sampe sentence to Google, which finally found me the new location of the article (on the web site of its original author).

Thus I edited this answer to include the new link (and the title of the article).

A short google search for the original URL on stackoverflow alone shows that the link was quite popular. So I started an editing spree, starting to replace each occurrence of this link with the new one. I always try to correct other issues with the post, as well, or course.

Have a look at my activity log to see how many edits I had to do for this (most entries with revised, starting at 16:15 UTC today).