ordinary ASCII text - still the winner
Friday, 28 January 2005 11:55 am![[personal profile]](https://www.dreamwidth.org/img/silk/identity/user.png){kind=small}
miriam_eDespite many attempts to supersede it, straight ASCII text continues to live on.
All manner of mutually incompatible wordprocessor formats have been developed in attempts to displace it. None has succeeded -- even supremely powerful Microsoft's own Word format, a strange format that is incompatible with everything else including later incarnations of itself, has made only limited inroads against the humble straight text format. PDF has tried very hard to replace simple text, but it remains a doomed, backward-looking format for people who never understood the electronic age and retain an atavistic fetish for paper.
Only one format has succeeded in becoming at least as popular as ordinary text, and that is HTML, thanks to the web. But HTML won't ever supplant text because instead of competing with it, HTML actually uses ordinary text in a cunning technique to display different fonts and styles and to insert images.
In recent years HTML has begun to invade email -- one of simple text's most strongly held territories. But in an odd turn of events HTML's attractive styles and flexibility are proving an impediment, as spammers now use HTML's hidden tags to evade spam filters, and embedded images to identify people who view email. This has caused many people to simply spam-filter all unknown email to the trash if it contains HTML tags. This may be resulting in a slow return to using simple, honest, text emails.
Text has also experienced a popular resurgence lately in the form of SMS messaging. It will probably always be the most economical way to transfer information from keyboard to screen.
In the end, if you want to guarantee that your audience can read your message then simple text is still the only 100% sure bet. After about half a century of computing this is still true, and perhaps will always be.
All manner of mutually incompatible wordprocessor formats have been developed in attempts to displace it. None has succeeded -- even supremely powerful Microsoft's own Word format, a strange format that is incompatible with everything else including later incarnations of itself, has made only limited inroads against the humble straight text format. PDF has tried very hard to replace simple text, but it remains a doomed, backward-looking format for people who never understood the electronic age and retain an atavistic fetish for paper.
Only one format has succeeded in becoming at least as popular as ordinary text, and that is HTML, thanks to the web. But HTML won't ever supplant text because instead of competing with it, HTML actually uses ordinary text in a cunning technique to display different fonts and styles and to insert images.
In recent years HTML has begun to invade email -- one of simple text's most strongly held territories. But in an odd turn of events HTML's attractive styles and flexibility are proving an impediment, as spammers now use HTML's hidden tags to evade spam filters, and embedded images to identify people who view email. This has caused many people to simply spam-filter all unknown email to the trash if it contains HTML tags. This may be resulting in a slow return to using simple, honest, text emails.
Text has also experienced a popular resurgence lately in the form of SMS messaging. It will probably always be the most economical way to transfer information from keyboard to screen.
In the end, if you want to guarantee that your audience can read your message then simple text is still the only 100% sure bet. After about half a century of computing this is still true, and perhaps will always be.
![[identity profile]](https://www.dreamwidth.org/img/silk/identity/openid.png){kind=small}
Re: How very true!
Date: 2005-02-02 06:49 pm (UTC)The exploit comes from the move in String and Pointer handling code from dealing with 7/8-bit ASCII characters to dealing with 16-bit Unicode characters, new font-processing logic, or the move for applications to allow changeable message sets to provide for localization.
Mishandled string buffers and font conversions and the like create exploitable buffer overflow vulnerabilities.
Programs like Internet Explorer handle arbitrary string input (from webpages/webservers made by just anyone) and have to do a lot of processing on fonts, character set encodings, the the basic text. Even a single C pointer mismatch could open a vulnerability, when dealing with programs as complex as IE
String buffer exploits can happen to any program that processes arbitrary inputs into buffers.
More info: http://securityfocus.com/library/category/29