Van Jacobson links are dead

[Qrbaker]

The Van Jacobson links to the video and slides on the Bloat Videos page are both dead. The video times out, and the slides were sadly shared via Google+, which was unceremoniously dumped in a ditch this April.

Given it is the primary source of the "fountain" analogy alluded to in the introduction page, I think its important to address this.

If someone has a backup of the video, could we look at getting it on a hosting site like YouTube or perhaps Peertube? Failing that, a written description of the fountain model would be a nice idea. If someone can explain it to me (I was trying to learn about it when I discovered this linkrot) I'd be happy to draft something.

[Qrbaker]

Holy cats, I managed to recover it! Using an old archive.org crawl I scraped the direct video ogv file link, then put that into the web archive and pulled it down. I now have the video file, so the question is now if we can upload it somewhere.

In the meantime, I'm working on a transcription so there is something of record.

[tohojo]

Quentin <[email protected]> writes:

Holy cats, I managed to recover it! Using an old [archive.org crawl](https://web.archive.org/web/20121127200746/http://recordings.conf.meetecho.com/Recordings/watch.jsp?recording=IETF84_TSVAREA&chapter=part_3) I scraped the direct video ogv file link, then put *that* into the web archive and pulled it down. I now have the video file, so the question is now if we can upload it somewhere.

Nice! How big is the file? We could just put it up on bufferbloat.net as a downloadable file?

In the meantime, I'm working on a transcription so there is **something** of record.

Great! Please do send a pull request with an update once you're done.

[dtaht]

thank you for caring. That was *the* core talk on bufferbloat. Quentin <[email protected]> writes:

…

The Van Jacobson links to the video and slides on this page: https://www.bufferbloat.net/projects/cerowrt/wiki/Bloat-videos/ Are both dead. The video times out, and the slides were sadly shared via Google+, which was unceremoniously dumped in a ditch this April. Given it is the primary source of the "fountain" analogy alluded to in the introduction page, I think its important to address this. If someone has a backup of the video, could we look at getting it on a hosting site like YouTube or perhaps Peertube? Failing that, a written description of the fountain model would be a nice idea. If someone can explain it to me (I was trying to learn about it when I discovered this linkrot) I'd be happy to draft something. — You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub, or unsubscribe.

[Qrbaker]

Nice! How big is the file? We could just put it up on bufferbloat.net as a downloadable file?

its 16.1 mb. I put a copy up on Archive.org: https://archive.org/details/video1_20191129, as well as just the audio, which I ran through a de-noiser and high-pass filter to try and remove the rumble/hiss: https://archive.org/details/bufferbloat_talk_audio

That was the core talk on bufferbloat.

That is what I gathered as well, which is why I'd be interested in helping update the overview page; maybe with a written fountain example to supersede the current traffic one.