Declare the Flickr API a National Historic Landmark | Hacker News Reader

Declare the Flickr API a National Historic Landmark

petitions.whitehouse.gov

10 pointsdavewiner18 comments

This is a pretty bizarre idea. You really want the government applying laws designed for historic buildings to software?

Even if this made any sense, when you declare something a landmark, you also make it nearly impossible to upgrade. You're saying that most of its value lies in preserving what was, not what will be.

Before you do the "waaaaah Flickr hasn't upgraded in a long time anyway"... http://www.flickr.com/services/developer/changelog/

They are hiring people to work on the API full-time, last I heard.

You get it. That's exactly the point.

Go ahead and make a new API, no problem, but don't break the apps we built over the last N years while Flickr was hibernating.

Would you be willing to accept the same strictures yourself, to software you have written? You are never allowed to sunset bad ideas or stop supporting them - at least, not until a government committee says so?

Yes, of course. And I did it with RSS 2.0, froze the format so we'd all have something to shoot for.

http://cyber.law.harvard.edu/rss/rss.html#roadmap

Perhaps I've misunderstood, but that does not seem comparable.

* You decided when the spec should be frozen. You're asking Flickr to give away control over their specs.

* Freezing a spec incurs no ongoing cost to you. You're asking Flickr to maintain current services indefinitely.

A comparable situation would be to put Userland under the control of some external entity, which had the right to impose obligations on you to (for instance) keep the data available in current formats forever.

Forget about the developers perspective and consider the users of the software that consume the API itself. Their applications will suddenly stop working and we cannot assume that they can obtain the upgraded version of their software (for various reasons).

That's right. Most of the development done with Flickr was finished a long time ago, and the apps are deployed and there's no one around to deal with deprecated APIs.

That's of course assuming that it's possible to do with a new API what you could do with the old one. We've seen with Twitter that's not necessarily true.

But thanks Dody for thinking of users. That's often pretty far down the priority list for developers, esp at big companies.

> It's a win-win. We get to keep building and Yahoo can say they're the first tech company to have one of its creations declared as an official national treasure.

Typically win-win situations don't involve threatening force (government) to get one side to comply. If Yahoo wants to deprecate the Flickr API, they are doing so because it is not a win for them.

Are developers willing to pay to keep the service running? If not, why should Yahoo be forced to?

Who said anything about threatening!

What an incredible windfall to have your API declared a historic landmark. What kind of marketing opportunity would that present. Please. Even a marketing idiot could figure that one out. :-)

Uh.

Dave, who cares if an API is a historic landmark or not? App developers certainly don't. Most devs view APIs primarily as functional objects (although yes there is an aesthetics to API design).

I think it's super weird for a government to mandate to another entity that it must maintain a piece of technology in perpetuity.

Besides, OSes aren't static anyway... Abandonware apps that depend on a static API are still going to bit-rot as new OS versions are released and fail to retain 100% backwards compatibility.

This just seems like a bad idea. I'm also still not clear on what the heart of the idea is. Is it to preserve the API itself as a work? Or is it to try and preserve the app ecosystem? Or is it to preserve a style of request?

I feel like if you really wanted to do this, the way the Internet deals with this is to come up with a common standard folks can adhere to, and then to lobby orgs to adopt that. Setting the Flickr API as a defacto standard (just for them) seems weird.

Of course that's why we should break everything -- because it seems "weird" to someone whose name is Know Theory. Right. :-)

Landmark personal and societal rights to social structures and databases is interesting psychotechnologically, in our real world where data rights, personal rights and property rights are so important.

Here's a blog post that explains...

http://threads2.scripting.com/2012/december/aNationalTreasur...

Dave

Even if this were possible, which it isn't. A landmark has to be a physical place/structure. It's a stupid idea.

Think different.

Still a stupid idea.

Neener neener.

This is a bad idea, and it makes absolutely no sense.

Reply on news.ycombinator.com