Henning Schulzrinne

@Delta Pandemonium at FRA DL107.

ChatGPT may not only replace the college essay, but also the faculty meeting.

Sign up for Post Prod! post.news/?r=EX2T1

A student-supported effort, drawing on Measuring Broadband America data, was able to provide helpful input to the new broadband label rules. fcc.gov/document/fcc-r…

@infinimatt The analysis was contributed by the Arial author, the math was done by CMR, the writing by Times Roman, and the Comic Sans author drew the pictures.

Instead of relying on author order to denote contribution to an academic paper, we should just use font size. 
“She is a 36pt author on this paper, and he is 6pt — I think it is clear who made the greater contribution!”

My take: youtube.com/watch?v=HXVZft…

The videos from the IBM Future of Networking series have been posted: youtube.com/playlist?list=…

My SIGCOMM talk is now at youtu.be/5lvXIqI_mQ4

@SteveBellovin @blakereid Why not assign digital object identifiers (DOIs)? The infrastructure exists, it's decentralized and the local naming part is relatively flexible. I have pursued, feebly, a version for regulatory decisions, which have the same problem (FCC Recd, anyone?).

@blakereid "The" blockchain is obviously a silly idea. But using a SHA2-256 hash, even truncated, might be interesting, and it would catch edits. For authorized edits, hash together the old hash with the new text. It's chained hashes, but without the needless consensus overhead.

@SteveBellovin Quark is very common in Germany - similar in some ways to Greek yoghurt. Great for baking, too (e.g., a less-unhealthy version of cheesecake).

We have two different varieties of this kind of cheese in the fridge. As I understand it, if I can find four more I can create all other types of cheese.

I just learned that Sandy Fraser, a pioneer in networking (en.wikipedia.org/wiki/Sandy_Fra…) and the former head of AT&T Labs Research, died last week.

I'm omitting the pseudo-keynote: "Here's my extended research talk combining my last three papers." I'm sure there are luminaries in the field where this type of talk holds the rapt attention of the audience. I always admire the self-confidence of these speakers.

Expansive keynotes (there must be a better term...) expand the view of a sub-discipline, e.g., by connecting the core or tools of a research area to new applications or techniques. They can often be inspiring, but this is getting harder since CS is now embedded everywhere.

Visionary keynotes are hard to make good. They often either seem like second-rate science fiction, with a particular aesthetic of trailing highlights, or they repeat the same buzzwords. AR! ML! Blockchain! Adding references to SDGs doesn't make a 6G keynote any better.

(Almost) every conference has a keynote - or a bunch. We usually think about the speaker, not what we want from a keynote: retrospective, visionary or expansive.

I like retrospective keynotes, where researchers or industry veterans share their perspective on a maturing sub-field or idea. What worked? What seemed like a good idea at the time, but wasn't? What were the surprises? Anything worth revisiting?

And here's the gigabit version - still ridiculous, but less so.

Verizon FiOS has weird pricing - gigabit is cheaper than 300 Mb/s, and both are prices that are way above industry norms, including $150 upgrade pricing. Their front page advertises $39 for 300 Mbps. Mistake or misleading?