Netscape Restores RSS DTD, Until July

Randall Bennett writes "RSS 0.91's DTD has been restored to it's rightful location on my.netscape.com, but it'll only stay there till July 1st, 2007. Then, Netscape will remove the DTD, which is loaded four million times each day. Devs, start your caching engines."

Why can't we just move it?

[Anonymous Coward]

Developers should take the opportunity to move to Atom. In the mean time we could use something as simple as round-robin DNS to share the load or have Mozilla, Google or the internet archive host it. It's a historical document and should reside at a permanent URI.

I don't get it

[Thansal]

I admit, I am not familiar enough with RSS. However this is a 2.3KB file that is not supposed to change. Why would developers NOT hardcode it into their RSS tools?

Re:Why can't we just move it?

[Anonymous Coward]

> Perhaps you mean "long-lived, mostly reliable URI".

Yes, and I have a couple of name suggestions for this, we could call it a "permanent URI" or "persistent URI".

Purl [purl.org] may be a good choice for this DTD.

Let's be Evil

[hackershandbook]

.. and I thought it was only Microsoft and Google that tried to "break the web" on purpose ....

Re:Why can't we just move it?

[Paulrothrock]

Kind of like Example.com [example.com]. That was set up in RFC-2606.

Technical vs. Emotional

[mmurphy000]

(I tried posting this as a reply to the blog posting, but I'm not getting the confirmation email, so I'll post it here)

From a purely technical standpoint, I agree with your assertion that, for well-baked files like RSS DTDs, clients should not be relying on a file hosted by an arbitrary service.

That being said, please understand that the emotional message you're sending is: "Don't rely on Netscape".

Why?

Back when RSS was first starting out, Netscape's documentation said to use Netscape URLs for the RSS DTDs. Witness this page [archive.org], published by Netscape, from late 2000:

Now, a shade over six years later, Netscape is saying "Oh, yeah, what we told you to do? Never mind. We're not supporting it any more."

If Netscape/AOL was shutting its doors, that'd be one thing. If the service in question was obviously onerous, that too would be understandable. Or, if Netscape told people "For the love of all that is holy, don't use our URLs for your DTD needs!" from the get-go (based on that document, you didn't), any such reliance would be our own fault.

But, because AOL does not want to serve up two static files, each of which is smaller than the "Netscape Reports" graphic on the netscape.com home page, Netscape is abandoning a service they told people to use.

So what are we to think about Netscape's current services and their long-term usability?

Re:Not really

[Bogtha]

I bet the bandwidth costs from attempted email delivery are huge even though there are no MX records and the server doesn't accept SMTP connections. In addition to spam harvesting, people like me have been using [e-mail] to satisfy email address requirements for years.

That's what the .invalid TLD is for, also defined in RFC 2606 [ietf.org].

".invalid" is intended for use in online construction of domain names that are sure to be invalid and which it is obvious at a glance are invalid.

URIs

[Anonymous Coward]

This is the perfect reason to use URIs hosted on p2p, rather than individual sites. It's going to be more and more of an issue, as RDF takes off.

Re: Because software evolves by mutation

[Vreejack]

That was insightful (hint to mods).

Now we need software that can breed sexually.

Or, more realistically, software that has a finer granularity and greater modularity so that the piece of ancient code that does this can be easily identified and swapped out, without needing to be understood by developers.
 

Re:Redirect

[naChoZ]

And any dev who codes his app to check a file like this every day instead of caching it client-side should be smacked oh-my-god-so-frickin-hard.

Ironic because Netscape is guilty of this poor practive themselves. I have an old sun u2 box that I recently revived. I had a copy of netscape messaging server/netscape enterprise server on it (used by the isp where I worked at the time). I wanted to archive some old mail off of it before I wiped the drive. I couldn't start it up because there were so many files containing references to http://developer.netscape.com/products/servers/ent erprise/dtds/nes-webapps_6_1.dtd [netscape.com] which of course doesn't even exist. Couldn't even start up until I replaced all references to that file with local file uri links.

Re:I don't get it

[rholliday]

I'm also not an expert, but from what I know about DTDs they are supposed to be referenced when the content should validate against them. For example:

<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.01//EN"
"http://www.w3.org/TR/html4/strict.dtd">

This is at the top of every Slashdot page. Should IE or FF break if the W3 were to remove that file? Certainly not. But should it be loaded and validated if possible? I believe so.

If any XML or RSS gurus want to correct me on this feel free.