Chris Dwan (he/him)
dwan.org
Chris Dwan (he/him)
@dwan.org
Freelance technologist supporting genomic and biomedical science. This is the tech nerd account. Somerville activism and local politics are at @somershade.bsky.social
Achievement unlocked: Keynote speaker.
April 3, 2025 at 1:07 PM
Most of the time, for me at least, the struggle and growth of writing is more important than the actual work product.

www.nytimes.com/2025/03/23/o...
Opinion | Sorry, Sam Altman, A.I. Is Not Good at Real Writing
When we let computers write our stories, we lose something essential.
www.nytimes.com
March 23, 2025 at 12:33 PM
Good thread on why AI for peer review (as described in this Nature piece) is a terrible idea.

bsky.app/profile/carl...
March 6, 2025 at 12:45 PM
For me, at least, it’s time to exit the phase of #AI hype where we all were putting raw ChatGPT* outputs in our presentations and talks.

It’s like time travel in fiction. If the story isn’t -about- time travel, then including time travel is just sloppy writing.

(*) or whatever #genAI tool.
December 16, 2024 at 1:32 PM
Top left, from the original Buddy’s Pizza just off 6 Mile in Detroit if I can get it.
New Haven, bottom right! Followed by NYC (although that’s not to my eyes a particularly good example photo). Then Detroit, and last, and by a good margin in my case, Chicago.

Note that St. Louis style is missing (not really a fan of their weird cheese, though).
You can only eat one for the rest of your life, which one you picking?
December 16, 2024 at 1:18 PM
The call for a pause in development of ‘mirror’ organisms is yet another in an accelerating series of cases in which we see the terrible risks posed by thoughtless use of new technologies, and yet are seemingly powerless to prevent harm.

1/?

www.nytimes.com/2024/12/12/s...
A ‘Second Tree of Life’ Could Wreak Havoc, Scientists Warn
Research on so-called mirror cells, which defy fundamental properties of living organisms, should be prohibited as too dangerous, biologists said.
www.nytimes.com
December 14, 2024 at 2:15 PM
Reposted by Chris Dwan (he/him)
If I can tell that your AI assistant is reading and (incorrectly) responding to my emails, I will supplement future messages with long, derailing passages from The Lord of the Rings or The Stand.
December 12, 2024 at 2:24 PM
Good morning! Here’s a quick thread of management tips, metaphors, and aphorisms I learned as a gardener.

1/?
December 4, 2024 at 2:20 PM
My only two interests in AWS reinvent are (a) who’s playing the shows and (b) which former partners will be The Sacrifice, paraded out to show the reactions on their dumb little faces as Lord Bezos cannibalizes his former ‘partners’ by launching native AWS services that destroy them.
December 3, 2024 at 3:16 AM
Reposted by Chris Dwan (he/him)
I'm getting lots of inquiries these days about finding a compbio job. This is the worst job market I've seen or heard of, and possibly the worst market since Bioinformatics became a field of its own.
December 2, 2024 at 8:08 PM
If, like me, you were thinking of putting your downloaded twitter archive up on ye olde webserver: Friendly heads up that the archive includes direct and group messages, and by default, those will be visible to the world.
November 25, 2024 at 1:05 PM
The paragraph below, taken from a job posting by a well funded startup, reads as a big red flag to me.

A couple of thoughts below.

1/?
November 19, 2024 at 4:23 PM
I’ve only had senior leaders say things like this to me a couple times in my career (‘if this was a good idea / important person, I’d have heard of it / know them already’).

In each case, it was a great indicator that it was time for me to move on.
And big-lab-group-think is a stifling influence. Last month I gave a talk at a local big pharma site and the director/host literally asked me “if [insert three big names of field] haven’t already thought of the idea you presented, how could it possibly be true/important?”
November 19, 2024 at 2:25 PM
Now that, for a brief moment in time, we've got a substantially less toxic platform, let's all resume live posting in real time about conferences.

Seriously, it's rad and I love reading this sort of in-the-moment coverage.
Shirley Liu, speaker at the Cancer R&D Conference in MA today. Describing using TRUST to scan for novel antibody sequences from RNAseq data. This stuff is normally tossed out early in analysis because it doesn't map to any reference genome!
November 19, 2024 at 4:22 AM
Seeing all these posts from #SC24 is giving me hope in a dark time.

Thanks to everybody who is letting their light shine through.
November 19, 2024 at 4:07 AM
Early #AI success stories in the life sciences (like openfold) were trained on massive, carefully curated datasets that took decades to create.

My guess is that there is no free lunch and that ‘garbage in, garbage out’ will hold true even in the age of AI.
November 16, 2024 at 6:30 PM
I wonder whether, for most people, reading and writing are going to follow the path of remembering phone numbers, doing math, and planning routes on maps.
November 16, 2024 at 12:52 PM
Early applications of generative AI make it clear, to me at least, that there are a -lot- of documents out there that nobody ever wanted to write and that nobody was ever going to read.
November 16, 2024 at 12:49 PM
I am absolutely thrilled (and more than a bit nervous!) to share that I will be joining Somerville indie rock band "Go!" on stage with my cello at the Crystal Ballroom, December 12, 2024 as part of the 2024 "Let's Rock Cancer!" benefit concert for the Dana-Farber Cancer Care Equity Program.
November 14, 2024 at 1:43 AM
Reposted by Chris Dwan (he/him)
Please join us at Portico Brewing Co. for the BCBB meetup, November 14, 6-8pm. 101 South Street, Somerville MA. Tacos are on us! RSVP here: www.meetup.com/boston-compu...
November 12, 2024 at 2:59 PM
November 12, 2024 at 2:06 PM
I’m pretty enthusiastic about #AI, but not -quite- to my mid-2016 levels of enthusiasm for #CLOUD yet.
November 11, 2024 at 2:57 PM