So many conferences are mid. Why was BSC 2025 so good?
68 points by bthompson
68 points by bthompson
I see another unusual thing about this conference that was not mentioned. From the conference website:
Where do I get tickets?Physical attendance will be invite-only. Tickets will not be publicly available. Invitees will receive an attendee guide with further information.
That might explain point 6, why “the caliber of attendees was very, very high”. Somehow, the three organizers knew who would be interesting to invite. Or perhaps they asked their first set of invitees to invite their friends. The website doesn’t describe the selection process.
Totally agree. I wonder if “conference” is the right term for such an event (it feels more like a “gathering of selected people”). Note that I don’t mean that such event isn’t valuable, just that it probably shouldn’t be compared to other public conferences.
It sounds like a retreat, especially given the scheduled rest days. Sounds great, and maybe there should be more, but obviously quite a bit different to a typical conference
How about “symposium”?
What about the Potsdam conference? (I dont think inviting the public is a prerequisite for a conference.)
aka Dagstuhl seminars :
The number of participants is limited to enable discussion and by the available housing capacity. The stay is full-board; participants are accommodated in the original house or in the modern annex, and have all their meals at the center. Seminars are usually held for a weekly period: participants arrive on Sunday evening and depart on Friday evening or Saturday morning. One or sometimes two seminars are held simultaneously with other small meetings.
I mentioned the RubyFringe and FutureRuby conferences elsewhere in this thread. They had public tickets, but were so obvious non-commercial, to the point of going for a counter culture feel, that the attendees self selected. Heck, it would probably have been tough to get a company to reimburse the tickets.
And even though virtually every monitor today has a refresh rate of at least 60 Hz, most conference recordings do not offer video in 60 FPS.
I’m digressing, but I think 60 FPS video is a bit overrated. Video “smoothness” is not entirely correlated with frame rate, because motion blur has a huge effect. Plenty of 25 FPS movies look smooth because they were shot with an aperture and shutter speed that create the correct amount of motion blur, and there are plenty of situations where slower frame rates are actually better since you need a long exposure when the scene is dark.
For some types of talks, I agree, but if the talk includes something like a technical demo with graphics, these are often in at least 60 FPS. I think it’s a shame when talk recording degrade the frame rate of these sections.
For example in Dennis Gustafsson’s talk it was cool to see him show off clips from the new engine. The clips had a high frame-rate in the amphitheatre, so by being in 60 FPS on YouTube, they at least still have a decent frame rate.
Ah, good point. I didn’t have this in mind. In that case videographers should be aware of that requirement, they may prefer different cameras, lenses and lights if 60FPS is a requirement. I didn’t mentionned it earlier but flickering lights are also a real source of headaches, in many places you can’t easily replace the light bulbs!
I think 60 FPS video is a bit overrated
I would upvote this 100x if I could.
There are two sides to this:
Point A, addressed to those to whom it matters:
You’re young and your senses are keen. Good for you. Be happy. It will not last and by the time you are middle-aged, it will be gone. Subjectively, when you are middle aged, it will have taken you a couple of years to get there.
The flipside of point A:
Most people aren’t young, and few people have keen senses. Most of us are relatively old, and the fraction of the world that’s old is increasing, rapidly. We can’t tell. Because we can’t tell, we don’t care. But we care very much about being able to do things like use our computers, because the more we physically can’t do unaided, the more we need our computers to do for us or help us to do.
This means: ensure you have a keyboard UI as well as a point-and-click one. Ensure it complies with keyboard UI standards. Yes those are a thing and they have been since before you were born. They are really important. They matter. Ensure your software is usable by people who can’t see, who can’t use gestures, who can’t use more than one finger, or who can’t use their hands at all. Ensure you have multiple ways to interact, with single button mouse only, with no button pointer only, with multiple buttons, and with no pointing device. Test this stuff. Disconnect your pointing device and learn to live without it. Switch to some alternative pointing device you don’t know well.
Even if you don’t care about others so much and you want it to look good…
One day you will be disabled, too.
Everyone ends up disabled, if they don’t die first.
Accessibility isn’t about helping others.
It’s about future YOU.
This comment is weird to me. While nothing you said is wrong per se, I don’t understand how it has anything to do with 60 FPS video. Would elderly or blind people have a better time with the videos if they were recorded at 30 FPS?
Ensure your software is usable by people who can’t see
In that context, I think it’s worth noting that of the pieces of “much needed software” mentioned in the post, the two that I could freely download and try (File Pilot and RAD Debugger) weren’t at all accessible with a screen reader. This seems to be a common problem with GUI tools produced by the Handmade-adjacent community.
This is the first time I’ve ever seen someone imply that recording video at 60 FPS is an accessibility problem. Why do you think that is the case? Which maximum FPS do you think is the limit before it becomes inaccessible to people in older generations?
I’m also struggling to understand the connection between the topic of video quality and the physical limitations that would prevent someone from using a mouse and/or keyboard. Could you expand on that a bit?
This is equally for @mort.
I’ve ever seen someone imply that recording video at 60 FPS is an accessibility problem
OK, obviously I failed to be clear enough.
That’s not the message here. I did not want to imply that, and it was not my intention. I messed up. Forget that, it’s a red herring.
My intention was this:
I can’t see 60fps video. It looks the same as 25fps to me. I can’t tell hiDPI screens from standard def. Right now my Firefox window is on an iMac’s 27” Retina display. To the immediate right is a 27” Thunderbolt display that is 1/4 of the resolution. Half the vertical, half the horizontal. Probably inferior refresh rate and colour range and contrast and so on too as it’s something like a decade older than the Mac.
I cannot tell. To me they look absolutely identical.
With my varifocal glasses on I have better than 20:20 vision and my colour vision is perfect. But I am old and have been short-sighted all my life. I’ve worn glasses all my waking life since I was 7.
The point I was trying and failing to make was not that there is anything wrong with 60 or 600 or 6000fps video. My point was: if anyone is able to perceive that video is 60fps and not half that then the ability to perceive that means that they are young and have keen senses and so they probably do not realise that other people cannot perceive such things… and that they themselves will not be able to perceive this things in 2 or 3 decades.
Every new bit of software that does things like support variable refresh rates, or HDR, or differential scaling on different displays… the tools that enable this, like Wayland and Mutter and GNOME… all of them are inferior in accessibility to the older display systems and desktops which they aim to replace.
I am not disabled, and my senses are fine, but I actively prefer older tools like X.org X11 and Xfce because their UI is better for me because I’m a keyboard user and rarely touch my mouse. I don’t use trackpads or gestures or touchscreens either. I actively dislike them.
There is a generational shift happening in tech right now and it’s driven by young people with good eyes who want GUIs that do things that I, at nearly 60 years old, am physically unable to perceive.
And a problem with these new tools is that they destroy established UI conventions, which their creators don’t even know exist.
This is a philosophical issue called Chesterton’s Fence. When you are building something new it is a natural response to not include old aspects of the thing you’re replacing that you didn’t use and didn’t need. Because you don’t use them, you feel they are unimportant, and can go.
But there is usually a reason those things are there and if you can’t see that then it does not means that there is no reason: it means that the reason is not obvious. So anyone who cares about 60fps video probably does not face issues that someone twice their age faces.
Merely caring about high-frame-rate hi-def high-dynamic-range and so on means you still have sensory sharpness that you are going to lose as you get older.
Nobody gives much of a damn about the wants or needs of boomers, but the things that help grumpy old folks in glasses are closely aligned with the needs of young people with sensory or motor problems. Everyone recognises a11y matters, but most people forget it. It’s an afterthought. Bolt something on later.
But there are core features of the UIs whose design inspired KDE and Budgie and Pantheon which the originals did with ease and which KDE and Budgie and Pantheon can’t do. They’ve been forgotten. As a grumpy old straight white guy, there’s no reason to care about what I want, and it’s increasingly clear that the designers of those desktops do not care.
I have found that I have allies in this: people with special a11y needs.
All I am saying is: learn what is there and why it’s there and then don’t rip it out.
I could not care less about being able to move windows around without them tearing. I can’t see it. Don’t care. But I care a great deal about being able to move windows around with keyboard controls and you know who else cares about that stuff? People who are physically unable to use a pointing device.
Nothing you wrote here is related to 60 FPS video recording.
Rendering to the display at 60 FPS has been supported by every UI toolkit and graphics pipeline since the ’90s, if not earlier.
The content and structure of your two comments here are alarming. I don’t know what time it is in your local timezone, but if you’ve been awake for an extended period then getting a good night’s rest may be of help.
To which, in general, I must ask: when you don’t understand something, but even judging solely by the length of the comment you didn’t understand, why on earth do you think it’s OK to cast aspersions at that person’s mental health? I mean, what the hell?
I also note that, despite the fact that I get torrents of downvotes for some of my comments, I can’t flag other people’s attacks upon me. That is not OK.
You still haven’t answered why providing live video recorded at 60 FPS would have any connection to input method accessibility, or age-related vision limitations, or contemporary graphics rendering pipelines.
If you want to write a blog post about how open-source developers ought to consider accessibility as a core feature when they abandon existing working projects to focus on mass-production of half-baked new projects then that might be a good thing to write. It sounds similar to existing criticism of GNOME’s development process, e.g. the well-known Cascade of Attention-Deficit Teenagers post.
But that topic has no connection with the comment you replied to or to the rest of the thread. As I’m sure you remember, 60 hz refresh rate has been the standard for LCD monitors since the very beginning of their mass-market adoption. XFree86 can render video at 60 FPS.
I did not cast aspersions on your mental health, unless you consider “hey man you sound like you’ve spent the last 36 hours breathing espresso while debugging xfwm crashes” to be a mental health diagnosis.
No, it is not related to the speed of the video, and it was written at about 3PM in the afternoon.
It is all about the ability to perceive high video framerates. It is about people and not video at all.
If older people can’t distinguish higher framerates but younger people can (is this really true?), then shouldn’t the accessibility cater towards higher framerates for younger people? I’m failing to see the disadvantage of higher framerates.
Most conferences are mid because they’re infested by sponsored and ad-adjacent content (e.g. developer advocates).
After that, the second most important reason why most conferences are mid is because conferences are inherently tools for community building, but in practice most conferences don’t cater to a real community. To be specific, a community in this context means a group of people that actively depends on each other for learning and pushing forward an ecosystem.
Zig has had so far good conferences because going to one means that you meet people that you learned from online, and whose projects you are probably using every week if not every day. At the conference those kinds of people give talks about things that they deem important to think about in order to enact concrete positive changes in the community.
The average conference doesn’t do much of that, and so it’s just you paying to get shown a bunch of ads and to do networking. The good moments I’ve had at conferences in the past all boil down to the point above (in my personal case that has been at Go and Python conferences).
Everything else is secondary (talk length, getting celebrity speakers, attendee selection, etc).
For context I’ve been a dev advocate in the past and I run software you can love, the closest thing to a Zig conference.
Zig also benefits from being under the mainstream radar (at least for now). People that are self-selecting into it rather than being told to. That results in people who are far more engaged.
conferences are inherently tools for community building, but in practice most conferences don’t cater to a real community. To be specific, a community in this context means a group of people that actively depends on each other for learning and pushing forward an ecosystem.
I completely agree, I don’t think I’ve seen a clearer crystallization of this before. It definitely applied at BSC too, although in terms of a dependency graph BSC was probably closer to a tree than a complete graph.
For example, while quite a few BSC attendees regulary use both File Pilot and Odin, File Pilot itself wasn’t written in Odin.
The Zig community strikes me as more interconnected. E.g. heaps of people (probably most?) who use the Zig compiler also use Ghostty, AND Ghostty itself is written in Zig, and (most?) of the core Zig team use Ghostty so the graph is more complete.
And continuing the graph metaphor, I can see how conferences which resemble a highly disconnected graph are disappointing.
Sorry to be the first person to point this out, but the photos and the speaker page have an uncomfortable samey-ness in who was invited and who was speaking. It gives the con’s touted “exclusivity” a bit of a dark edge to me.
I agree with @reezer about:
The other side is politics creeping in heavily. And while it usually aligns with my political views I don’t really like seeing that so much. While giving broad “human rights” statements, if it fits the context (eg. around accessibility, and simply access to information and such) make sense, or that you shouldn’t hard people obviously, when there is constant pointing out of a developer being some minority, some sexual orientation, gender, some ethnicity for essentially no reason at all it becomes distracting.
… that it’s distracting when a conference insists on patting itself on the back constantly for its inclusiveness (and, honestly, comes off as a bit defensive). But maybe that’s better than the alternative presented here, which looks very lonely and alienating, like a frat party.
FWIW, I noticed this too and almost commented about it earlier. The photos in the blog post look like a lot of dudes duding it up dudely.
In the game industry for many years there was a prestigious invite-only conference with a similar vibe (plus a strong culture of heavy drinking), and people who were invited reminisce about it a lot. But I get the impression it also pushed some of the talented people who Looked Wrong out of the patronage networks and reinforced the weird artistic conservatism of the industry as a whole.
samey-ness in who was invited and who was speaking.
California has a large latino population. But, when I worked at “big company” in California, it always seemed like the only latinos working there were in the cafeteria or facilities and not developing software. At the end of the day, representation matters. If young people can’t see themselves in a particular position, it can be intimidating. I don’t know why this is hard for some people to acknowledge.
In regard to this conference, there was a 2.5 hour talk. That is just excessive and unnecessary. If the point can’t be made in 30-40 minutes, the talk needs editing. They could have limited the time and had at least three more talks from other speakers.
This whole conference is a reaction to a different conference becoming more inclusive and lacking the depth they wanted. So your vibe check is pretty spot on.
Most of us have attended our fair share of mid to disappointing conferences
Agree a lot. Public massive conferences nowadays are always disappointing, only existing with the sole purpose of boosting organizer egos, getting extra income from sponsors or being giant publicity stunts.
The same applies online, with every public corner of the internet worsening, whether the mods want it or not.
Invite-only communities that produce public content is the way to go in times of scrapers, bots and grifters ruining everything.
I always like it when conferences post their talks online and this conference was all about that. I generally prefer to watch conference talks on Youtube these days.
I’ve indeed been to so many conference talks that were very clearly not all that interesting 5 minutes in and it’s socially awkward to walk out. Something that’s easier to remedy when it’s an online video.
A trend that is coming up more and more is talks that are 20 minutes with no QA. It comes from the JS world and I really don’t like it.
I wish there was a centralized platform/channel that aggregates conference talks. I would pay money for that.
I wish there was a centralized platform/channel that aggregates conference talks. I would pay money for that.
I think you’re onto something. I’d love for there to be an ad-free platform where I could browse by speakers and topics.
Some speakers keep lists of their talks on the websites (e.g. Andrew Kelley) while other people have unofficial talk achives (e.g. Bryan Cantrill) but the fact that there’s no database means many can be difficult to even discover.
I wish there was a centralized platform/channel that aggregates conference talks. I would pay money for that.
For Ruby, we have https://rubyevents.org/.
I started to avoid going to conferences. While more niche stuff is typically fine, it seems a lot of the bigger stuff “lived long enough to be commercialized” too much. Everyone is trying to sell you stuff in one way or another. Things shifted from “Hey, this is a cool tool” to “Use this tool, because it’s will do everything and has no downsides”. But of course it still needs support which you can get from this and that company.
The other side is politics creeping in heavily. And while it usually aligns with my political views I don’t really like seeing that so much. While giving broad “human rights” statements, if it fits the context (eg. around accessibility, and simply access to information and such) make sense, or that you shouldn’t hard people obviously, when there is constant pointing out of a developer being some minority, some sexual orientation, gender, some ethnicity for essentially no reason at all it becomes distracting. Don’t get me wrong, perfectly fine for people to do that in general, if they feel the need to, but when it’s essentially unrelated to the topic of the talk it gets that feeling of an animated gif next to a text. It’s not necessarily making the content bad and it’s not bad on its own at all, it’s just making it hard to focus on the topic.
It’s essentially the feeling when you contact some sales person of some product to learn whether it would fit your use case, but the constant buzzwords and the general “I want to sell you something” attitude is just annoying. Oftentimes you just want to know the price and access to docs. And in conferences when the company someone works for or some personal trait overshadows the content of the talk it’s not what I’d call a good conference.
The best talks are usually the ones where I forget what the name of the person that did it was, but something I act on frequently, and remember random details.
A similar thing happens when every second slide is a meme or a video - unless of course these are somehow the content.
That said, I don’t want to tell anyone how to do their talks. Just like I won’t tell people how to do their personal websites. I think it’s great if you do them in first place. It’s just that I think maybe it’s worthwhile to trust in the content of the talk a bit more. If you want to make it sharp and to the point, don’t fill it with cat pics. Again, when done once or twice or really as an aside theme that’s one thing, but lately those things that are aside start distracting from the main content so much. It’s like the web without ad blockers.
I read it. (Estimated reading time: 9 minutes, it says. I thought about, oh, 90 seconds, myself. Whatever.)
For anyone who goes to SF conventions: it was a small tech conference that was run like a small, lightly-programmed SF con.
If you’re a con-goer, that’s probably enough: you get it.
For non-con-goers: this is why I go to SF cons.
Yes, tech cons should be more like SF cons. Go to SF cons, talk to conrunners, find out how it works.
I’ve been to some pretty small SF cons and this sounds smaller and more tightly focused than that. more like a writer’s workshop/retreat if you want an analogy from the sff world.
A similar great conference (I felt, as a listener and a two-bit presenter) was last November’s dtrace.conf(24). Some related features to BSC would be that it was relatively small (though as an online conference of only a few hours, this is more a matter of vibes), had a bunch of really interesting invited(?) talks on the focal topic at hand, and as an unconference didn’t put explicit time limits on speakers, instead mostly allowing things to run their natural course, without forgetting about Q&A.
I personally tried to gauge interest (from speakers I’d love to hear talk more) in an online unconference on data-oriented design but received only crickets in response :) Perhaps a BSC-like format would have more pull?
While some of the particulars are different, this reminds me of the RubyFringe and FutureRuby conferences put on by Pete Forde years ago in Toronto. It was 100% a function of the personalities behind it. Full day of interesting presentations in a single track, full evening of hanging out together at amazing cool quirky places. Their measure of success was breaking even. No corporate sponsors. It was pure and sweet, in a way.
Is this a game dev focused conference? Lots of the speakers and topics are game related.
It isn’t but in game-dev (at least historically) there is(/was) a big focus on performance targets through frame times ( 60 fps means 16ms of time to do stuff in per frame), limited hardware (targeting consoles, especiialy old ones) and on the PC side targetting inclusivity despite hardware (targetting older hardware).
This lead to a group of developers who take/(took) performance seriously, and thus also know what is feasible. Thus when they look more generally at software (that isn’t video games) they see programs that should not be slow, becuase they know they can (and do in fact do) much more computation in the same time.
The conference premise is basically “Wirth’s law” - software is getting slower more rapidly than hardware is becoming faster. It is therefore natural that the primary group of people in the community come from the game dev group.
