A Redundant Array of Independent Dans
We would like to invite everyone to join us 28 May-1 June 2024 for the twentieth year of BSDCan!
It’s a new registration system, so we’re rolling it out in phases. Speakers have been asked to register, but you’re next. Don’t wait, register now.
(by Pamela Mosiejczuk and Michael W Lucas)
We just sent out submission acceptances and rejections. We’re months out from the conference and everything can change, but if everything works out we’re gonna have a great con. So let’s talk about rejection.
Conferences reject submissions for reasons. Submitters feel that those talks get rejected for completely different reasons. It’s not that unaccepted talks are bad. Conferences generally weed out a very few submissions where the abstract was not of a quality/completeness that we could actually judge, sure, but these committees aren’t gauging your worth like an academic board. We’re tasked with creating a lineup like a programming block for television.
After years of working with poster sessions, colloquium, and conference scheduling, especially for BSD conferences, here are the common reasons why we turn down a submission.
Note that this list does not include “your proposal was bad and you should feel bad.”
People are reluctant to re-propose the same talk again. Don’t be? Maybe people filled up the fuzzy bunnies track really early the first year you submitted and yours was simply one of the sacrificial ones because a few of the others fit together especially nicely thematically. The next year, the masses might be on to slithery snakes and there will be a prime space for your etymology of fuzzy bunnies talk, which was always interesting, there was nothing wrong with the bunnies!
People forget that we are generally not gauging quality at all, we’re rating things like “is this clear for this audience?” and “can this talk be a good way to get people to meet during the discussion and be able to continue to chat into lunch?”
There are certainly ways to slamdunk a submission to get it noticed. An abstract where we can imagine the precise audience who will be interested, which makes it obvious the level of experience required, talks clearly about the topics to cover (and sounds reasonable for the timeframe and experience of the presenter) and explains how the audience will be able to use what they’ve learned for their own purposes? That is immediately noticed. One that simply presents a topic without much context requires that both the committee and the audience know exactly what you are on about and can identify the talk’s purpose and end results on their own. If you want to talk about filaments in light bulbs produced between 1982-1995 we NEED to understand what is interesting about that. We believe you, but you only get a paragraph or two, is it there? Showmanship does matter, but this detail is why we can get excited about your topic.
Sometimes it’s just not exciting, it’s informational. Then we need to have someone around who understands its value. Teams tend to have both specialists and generalists doing these reviews. I’m a generalist and will alert on talks providing interesting looks into history, social topics, future planning, etc. Someone else might be the light bulb person. If we don’t actually represent that specialty to judge well? That’s OUR shortcoming, not yours.
This process is a balancing act where a handful of people try to stack a bunch of blocks in such a way that we’ll hit a nice variety of topics and provide an interesting broad cross-section of what’s going on in a particular world without letting the blocks fall because we didn’t provide a sufficiently varied mix of popular and niche, this topic and that, intro and advanced, rote technical and something-new.
Also, we all want to hear these sample talks about bunnies and light bulbs. But not at BSDCan.
Also also, you should know that the people who handle BSDCan’s money have no say in which papers get accepted. BSDCan has never sold talk slots. (I would say “never will,” but a million dollars up front and a generous trust fund for every attendee might change our minds.)
With that: here is the tentative list of talks for the 20th annual BSDCan! (Subject to change, slippery when wet, use as directed, do not taunt Happy Fun Ball.)
Tutorials
Managing OpenBSD Networks with NSH — Tom Smyth
Network Management with the OpenBSD Packet Filter Toolset — Peter Hansteen
BGP 101 — Massimiliano Stucchi (AS58280.net)
Run Your Own Email Server — Michael Lucas (BSDCan ConCom)
BoF
ZFS BoF — Allan Jude (Klara Inc.)
Implementing Routing Domains on an OpenBSD workstation for use with WireGaurd — Josh Grosse
Contributing to FreeBSD
Contributing to NetBSD
Talks
Supporting Business IT and network needs with OpenBSD and NSH — Tom Smyth
HardenedBSD 2024 State of the Hardened Union: A Decade of Hardened Bits — Shawn Webb (The HardenedBSD Project / The HardenedBSD Foundation)
A Journey Into BSD and Standards: BSD and POSIX — Katie McMillan (Government of Canada)
The FreeBSD and Windows Environments — Michael Dexter (Call For Testing)
NetBSD Subfiles — Elijah Sherwood (Western Washington University)
The State of Email — Michael Lucas (BSDCan ConCom)
Zelta: A Safe and Powerful Approach to ZFS Replication — Daniel Bell (Bell Tower Integration)
Alamosa: A Tiered Disk Block Cache for NetBSD — Kira Ash
20 Years of NYC*BUG, and Can We Handle 20 More? — George Rosamond (NYC*BUG)
Supporting FreeBSD in the Field — Allan Jude (Klara Inc.)
LLDB FreeBSD Kernel Module Improvement — ShengYi Hung (National Taiwan Normal University)
Summa Tetraodontidae: Thomas Aquinas Explores OpenBSD’s Medieval Orderliness — Corey Stephan (University of St. Thomas (Houston, Texas))
Address space reservations: Re-thinking address space management for pointer provenance — Brooks Davis (SRI International)
NetBSD on RISC-V – It Finally Runs NetBSD — Dylan Eskew (Western Washington University)
Running your own network using BGP, OSPF and IS-IS on the BSDs — Massimiliano Stucchi (AS58280.net)
Making NetBSD as a fast(er) booting microvm — Emile Heitor (NetBSD)
The Accidental Release Engineer — Colin Percival (Tarsnap Backup Inc.)
Why fsync() on OpenZFS can’t fail, and what happens when it does — Rob Norris (Klara)
Supporting a development lab with FreeBSD — Chuck Tuffli (FreeBSD)
DJ-BSD: DJing and music production in FreeBSD — Charlie Li
Encouraging and enabling SMEs (small to medium enterprises) to contribute to BSD development — Tom Smyth
A userspace NVMe over Fabrics Implementation — John Baldwin (FreeBSD Project)
Calling the BATMAN: Free Networks on FreeBSD — Aymeric Wibo
Improving Haskell Development Experience on FreeBSD — Daniel Lovasko
How to get started hacking NetBSD — Taylor Campbell (The NetBSD Foundation)
quiz: tiny VMs for kernel development — Rob Norris (Klara)
Contributing to FreeBSD via Github — Warner Losh (Netflix)
Towards a Robust FreeBSD-Based Cloud: Porting OpenStack Components — Chih-Hsin Chang
USB Debug Capability on FreeBSD, Revised — Hiroki Sato (Tokyo Institute of Technology)
Hot cross builds: cross-compilation in pkgsrc — Taylor Campbell (The NetBSD Foundation)
FreeBSD as the backbone of a vaccine/medication refrigerator monitoring system — Phillip Vuchetich (Arxsine Inc.)
FreeBSD at 30 Years: Its Secrets to Success — Marshall Kirk McKusick (McKusick Consultancy)
Full Disclosure! BSDCan Finances
The CFP for the 20th Annual BSDCan has ended and the program committee is busily reviewing your submissions. I’m not only the organizing team chair, I’m also on that papers committee, and all I’ll say is that we got some good stuff this year.
Thanks to the efforts of Allan Jude and Colin Percival, BSDCan now has a new business organization and a bank account. Sponsor support is starting to come in and prospective sponsors often rightfully ask us how we spend their contributions. While numbers increase every year, here are the rough figures from 2023 (in Canadian dollars, because Canada).
Total: $74,000
One of the things that makes BSD conferences special is that we pay for speaker accommodations and travel. This breakdown shows why we need sponsors. 2024 will cost more than this.
Note that this does NOT include the FreeBSD DevSummit. The FreeBSD Foundation pays those expenses. Some speakers attend the DevSummit as well as the conference, but that doesn’t increase the flight cost so we’re delighted to help out there. Some speakers volunteer to cover their travel costs and we sincerely appreciate it. If your employer will cover your travel and accommodation, we’d be thrilled to list them as a travel sponsor. If you want to host a similar DevSummit or mini event, reach out immediately so we can get you a room. My next post will be too late.
BSDCan registration fees haven’t changed for years and we pride ourselves on keeping costs low. We could pay for the entire conference by raising the admission to $750 or so, but BSDCan’s traditional motto is Bringing BSD Together. We want to welcome younger folks who will become tomorrow’s contributors. Instead, we’ve slightly increased admission for regular attendees. If you work for a large company that has a serious training budget, please sign up as a corporate attendee. We’re being pessimistic and thinking we can raise $30,000 that way. We’re also planning to charge for the closing social, but there’s a possibility that a per-head fee will keep some folks away. Who knows how that will work out?
That leaves $44,000. Our sponsorship goal is $50,000 to give us some leeway for increased costs. We’ve been promised a good chunk of that so far, but only about $8,000 has arrived in our account and some of that money has already gone—reserving rooms means paying deposits.
For BSDCan to succeed, we must make pessimistic assumptions. We might have more attendees. We certainly hope to have more attendees. Any funds raised over the goal will go to replenishing BSDCan’s financial reserves for 2025 and beyond.
If you work for a company that uses BSD, they’re getting a heck of a deal. I’d encourage you to ask your employer to sponsor us. They’ll want to know why. Here are some things you can use to persuade them to support us.
Why Support BSDCan?
BSDCan’s traditional motto is Bringing BSD Together and we do that in many ways. For twenty years we have brought developers together to discussion technical matters, helped users learn from authoritative presentations and tutorials, and brought people of every level of experience to the hands-down best BSD hallway track in the world. No matter where you’re from or what your level of BSD experience is, you are welcome at BSDCan.
BSD Unix and its projects are about building quality infrastructure. We would not have the Internet as we know it without BSD, nor many of the most popular appliances, phones, and web sites. This reach is possible thanks to our commitment to sharing our work without charge and without restrictions. This permissiveness spurns innovation and has done so for nearly half a century.
What is on the BSDCan program?
Users and developers will discuss real-world problems and resolve edge cases to enhance system reliability, performance, and usefulness. Tutorials expand users’ skills while various projects will host their own meetings. Well-known BSD technologies represented at BSDCan include OpenSSH, OpenZFS, Packet Filter, IPFW, OpenPAM, Jails, bhyve, IPFW, DTrace, LibreSSL, and Clang/LLVM among many more.
Who is the target audience?
The target audience includes everyone from potential BSD users up to advanced developers. Tracks include Development, Systems Administration, and real-world Experiences.
Our core constituency includes:
We are specifically reaching out for two other groups:
How does your sponsorship help?
BSDCan has always kept its ticket prices low to allow the broadest demographic of community members to attend. We also pride ourselves on funding the travel and accommodations of our presenters to further remove barriers to participation. Your sponsorship goes directly to our venue and travel costs, plus luxuries at your discretion such as the social event.
Visit our Sponsorship page (https://www.bsdcan.org/2024/sponsorship.php) to learn more and we hope to see you there!
Michael W Lucas – for the BSDCan organizers
We need you.
While sponsorships have started to arrive, we have several potential large sponsors and are honing our pitch to them. They want to know how BSDCan has changed both people and organizations. I don’t know how often I’ve seen some bewildered first-time attendee asking someone about their problem and be led to the person who created the code responsible. I’ve lost count of how many times I’ve seen that same newcomer standing in a tight group with a handful of other users and developers discussing exactly how the software should function. Many times, I have seen that same “bewildered newcomer” return the next year to present on their solution for the problem.
I know many people who found employment via BSDCan.
I know of many companies that decided to use BSD software, came to BSDCan seeking guidance, and learned we could solve not only the problems they knew about but ones they hadn’t yet considered.
Potential sponsors want these testimonials.
If you or your organization are one of them, and you’re willing to go on the record saying that BSDCan changed your life for the better, please send us a brief testimonial. You can comment below (please sign your real name, not “DeathDaemon1999” or whatever your handle is), or email it to me at mwl at mwl dot io.
In other news: all web sites are painful, but sites from 2004 are differently painful.
We now have a functional 2024 web site on the new team’s infrastructure. It’s the same old BSDCan web site we all know and, supposedly, love. I’m not knocking Dan; this is how it was done, back in the day. Bringing this site up has really driven home that the operations team includes zero front end people. We would really like to archive the historical sites and start over, but have no good options for doing so. (Yes, we could slam up WordPress and get something running, but I said good options.) We need something that can easily be replicated and archived year after year, but works well on mobile devices. If you’re a web designer who regularly attends BSDCan and want to get involved, do drop Lucas a note at mwl at mwl dot io.
The CFP was supposed to be live 1 December. You’ve probably noticed it isn’t. David is working hard, and we should have a submission system ready soon. When it’s ready you’ll see it here, and on the announcements mailing list and the various social media sites.
At a vendor meeting last month, Dexter spoke with several prospective sponsors. We’ve implemented their suggestions, and simplified the sponsorship tiers and created a handy brief explaining why people should sponsor BSDCan. We’re approaching potential sponsors hat in hand, but you can do that same. If your employer uses any BSD-adjacent technology, such as OpenSSH or OpenZFS, our conference aligns with your business interests. While I’d love to have a single $50,000 sponsor, ten $5,000 sponsors is far more sustainable. Getting your company immortalized in the conference videos is the kind of publicity money can’t buy—except it can. If you’d like to help us restart without full-on sponsoring us, we’re taking donations via Paypal at funding@bsdcan.org.
BSDCan is now an Ontario non-profit corporation. This simplifies a whole bunch of stuff, at the cost of Allan and Colin doing extra paperwork. Everybody but Colin and Allan are fine with this, and those two don’t mind too much.
Things are coming together. I expect this month to have an extra blog post once the CFP launches.
(MWL, for the ops team)
Actual news this month, including: Dates! Donations! Dinosaurs!
No. Wait. No dinosaurs. Sorry, my bad, carry on.
The bsdcan.org web site has a splash screen giving the 2024 dates. Why a splash screen? We’re still looking to get the web site moved from Dan’s control to ours. Dan is being fully cooperative, but the computers are being jerks. We have hopes for this week, but we must get the date out this week. (Why this week? FreeBSD is having a vendor summit this weekend and our fine, fearless, fulminous sponsorship coordinator will be there shaking hands and smiling and holding out the hat. Having actual dates will grant his efforts gravitas. (If any other BSD project invites a bunch of vendors to a meeting, we would appreciate a heads-up so we can dispatch Mister Dexter. (Better him than me, that’s all I can say.)))
Speaking of money, quite a few folks have offered to send us a few bucks to help us restart. The email address funding@bsdcan.org is now live at Paypal. Your nickels and dimes will help us grease the rails into 2024.
We’ve also decided on a greater amount of financial transparency. Our goal is for BSDCan to develop sufficient resources to carry us through another bad year. We won’t achieve that in one year, or two. Maaaybe five, if we’re lucky. But a couple slides at the closing session saying “This is about what we spent, this is about what came in, here’s where we are, thank you” will go a long way.
We’ve had a serious discussion on the mask policy, and the consensus is it stands for 2024. None of us like it. We all eagerly look forward to the day it won’t be needed. Some folks live in communities that have abandoned masks except for medical reasons, while other communities still wear them rigorously. But many of us are especially vulnerable. We are biased towards safety, and while we can’t control the hallways where students walk around we can control the event rooms.
We also had another discussion around ancillary events. We want to remind folks that we’d like to welcome any projects that want to host adjacent meetings. If you’d like to plan such an event, please let us know as soon as possible. Adam is still running under the Big Giant Lock and needs time to prepare.
So. What’s your favorite dinosaur?
A short blog, because we’re in the slow time.
Over the last twenty years, Dan built up a bunch of BSDCan technology infrastructure. Andrew is busily scooping that out of his hands and migrating it to servers under his control. Once the web site is under the current team’s control, we’ll launch the 2024 edition and announce the 2024 dates. We won’t be using all of Dan’s infrastructure. Paper submission and conference registration systems are solved problems, and re-solving them is neither educational nor interesting.
Colin and Allan are still setting up the Canadian business to handle the money. Paperwork, government, all that translates to “hurry up and wait.” They’re pretty much settled on non-profit. A Canadian non-profit needs three directors, so Peter Hansteen got volunteered. We specifically wanted a director who was not part of FreeBSD. Peter’s main role will be to say, “Uh, guys, what about NetBSD?” (Allan and Colin will certainly do their best to be inclusive, but they’re unabashed FreeBSD folks.)
We’re also discussing financial transparency. Ideally, we’ll provide a subject-to-much-revision rough cut during the closing session, and a more complete report after the conference closes. We will not detail individual receipts for surge suppressors and whiteboard markers because someone will nitpick that we didn’t need them and even if we did we could have gotten them cheaper at https://FellOffATruck.crime — but on the other pitchfork tine, if we’re asking for money we must document that we’re not squandering it. Like any venture, our goal is to build up sufficient cushion to carry BSDCan through lean years.
We plan to get the the CFP out at the beginning of December and close it 15 January. That’ll let us announce accepted papers in February and let folks start booking travel.
Migrating two decades of accumulated tech is ugly, thankless work. If you see Andrew, buy him a drink.
Dan Langille ran BSDCan for nineteen years, building it into one of the keystones of the BSD community. This was a stunning feat of service. Dan has chosen to step down, handing the responsibility over to a group of volunteers.
The operations team meets monthly. We have a decent grasp on the mechanicals of running an event and have divided up responsibilities. Adam has opened negotiations with the university. Colin and Allan are working out the details of opening a Canadian business to handle money. (Our gut reaction is this should be a non-profit, but that’s a matter to discuss with accountants. Both Colin and Allan run Canadian businesses, so we’re leaving it to them.) Lucas is busily deflecting all actual work onto other people.
The critical issue today is money. (Canadian dollars, because Canada.) BSDCan isn’t quite broke. The 2023 event, with its dearth of sponsors, mostly drained our resources. Don’t get me wrong, we love the sponsors who showed up for us! But we needed more. We have a few thousand bucks for seed money, and that’s it.
BSDCan needs enough money to pay for:
We would like enough money to pay for:
How much money does BSDCan need?
About $80,000. We might get away with less, but prices are going up so it might be more.
How are we going to get it?
First, admission has remained $195 for many years even though supplier prices have increased. We want to raise it to $250. We had 130 paying attendees in 2023, which was pretty good for the first in-person event in four years. 150 attendees would bring in $37,500. That leaves us over $40,000 short.
Second: when we have a Canadian business organization and bank account, we’ll have a donate button. That’ll certainly help.
We’ve set a goal of raising $50,000 from sponsors. That should let us tolerate random price increases from airlines and the University of Ottawa. I requested that any extra go towards paying down my gelato bill, but the committee is insistent that excess funds be retained for improving BSDCan. We must also build a cushion for future years.
Michael Dexter ( dexter@bsdfund.org ) is our sponsorship coordinator, partly because he’s good at those things but mainly because he’s the only one of us who owns a suit. He looked at the issue and said, “The real problem is that BSD geeks are actively terrible at raising money. We should explore ways to reduce the stress of funding all of the BSD events.” You can find his initial efforts at bsdfund.org while we work out the BSDCan infrastructure.
BSDCan’s sponsorship options accreted over the decades. If you wanted to sponsor T-shirts, Dan would make up a “T-Shirt Sponsor” category and take your check with a smile. Some of those options don’t make as much sense today. Other options have not increased in price since the naughties. Dexter went through them, boiled them down, and built a rationalized table of sponsorship options. He’s starting to knock on doors right now. The front page of this site has a funding thermometer to track his progress.
Without sponsors, BSDCan doesn’t happen. We could raise admission to $600 and cover everything directly from your pockets, but that’s a dramatic and hideous change. BSD software runs some of the biggest industries in the world, and we anticipate that a handful of them are still willing to give back. If that describes your organization, please check out the sponsorship options. Dexter would love to hear from you.
Twenty years of BSDCan has made the event something of a sacred trust. Everyone is working together to continue it into the future. That starts with money.
Bringing BSD Together
Running BSDCan consumed Dan’s bandwidth, leaving him with zero interest in publishing financial reports. That left space for confusion. Let me categorically state: BSDCan does not pay for project devsummits flanking the con.
The FreeBSD Foundation is traditionally a conference sponsor. The FF also pays for the FreeBSD devsummit. They also have a separate program to sponsor travel for folks who want to attend. Yes, they leverage BSDCan paying for speaker flights. They should. Other projects should do the same.
BSDCan’s motto is bringing BSD together. Having an operations team rather than a Single Point of Dan will make that easier. We would happily reserve space for related projects such as bhyvecon or NetBSD or whoever, and give such projects a couple minutes at the closing to say how things are going. Video coordinator Patrick McEvoy has volunteered to stream such events, provided that they happen right next to the main con and he has enough cameras and staff. We will be asking all such devsummits to pay room fees in advance. Total expenses wobble with the number of lunches needed, so we’ll also have to settle up afterwards.
If you want to hold such a meeting, watch this space. We’ll figure out a process soon.
One day I’d like to see several related groups meeting for a couple days before BSDCan and all gathering for lunch. If I let my imagination soar, some of those lunches would not involve sandwiches.
We’re working to convert BSDCan to a team approach, and want to document our efforts–both so that we can keep our facts straight, and so that the community knows what’s going on.
We’ll make a post after each monthly meeting. We want to be very open, except for information that will make people take ill-advised actions. For example, we have picked the best available dates for BSDCan 2024 but have not yet gotten agreement from the university. Publishing those dates would inspire some of you to schedule your lives around them and then get annoyed when the university rejected them. (Not you. You’re sensible. Other people.)