Googlet

The Google Gadgets are a cute idea, they allow to encapsulate Web forms into pieces of XML, which then can be added to personal Google start pages and similar services.

A bit like the old favlet/bookmarklet concept based on javascript-URLs, but with more features like user preferences. I'm not yet sure where they store user preferences, on their servers or in cookies.

The name add.gif for the link icon isn't very intuitive, I renamed my copies to googlet.gif. My second experiment after the xyzzy "custom search engine" is a LEO dict search form, but so far Google claims that it's unavailable or empty when I try to add it.

sitemap.cmd

For users of sitemap.cmd, a REXX script generating sitemaps, the improved sitemap specification doesn't require an update.

The location of the sitemap can now be given in a robots.txt file, see their protocol page, example:

 Sitemap: http://www.xyzzy.claranet.de/sitemap.xml 

Any services above finding the sitemap like Google's webmaster tools still require some form of submission proving that the submitter has write access on the submitted site.

Syndic8

History, this isn't my first attempt with a "mail2blog" service, for two years mailbucket.org did what I wanted.

Unfortunately guessing the submit address for a mailbucket feed was simple, and the spammers found it. Other services I tested (including blogger) had no "mail2blog" feature OR required a version of SSL not supported by my browser(s) OR required a version of JavaScript not supported by my browser(s). Therefore I was forced to drop my old (pseudo-) blog redirecting to an error page without proper HTTP error code.

Now after it works again I found that Syndic8 never gave up to poll the feed URL for its FeedID 62125.

Validation issues

The Blogger help pages are messy, many links are broken, SSL login fails miserably with old browsers, the feedback link without login doesn't work, etc. If you're looking for a blog hoster try to find a better service. I want the "blog-by-mail" feature available here, so it's my own fault.

Known issue, the application/atom+xml media type for the atom feed apparently degenerates into text/html after the first access. This might be a cache issue, resulting in a warning by the W3C validator:

HTML test

Here's my OpenSPF page, it offers a customized search engine provided by Google.