Minutes of the SIPB Meeting of 2012-02-20 The meeting was called to order by dwilson at 19:30. In attendance were Voting members: ezyang adehnert phurst kaduk achernya lapentab bbaren ine | cesium Associate members: | jhawk mitchb zhangc kcr klee Prospectives: dlaw madars Guests: Administrivia ============= dwilson: Today we will be electing the SIPB executive committee. kaduk: coleco-sidewinder does not have AFS installed on the OS X Lion partition. I wonder if anyone knows why it wasn't installed. Otherwise, I will try to convince someone to install it. Project Updates =============== achernya: Linerva had issues yesterday, where users weren't able to log in over SSH. We got a root shell by accessing it directly through console. Our first thought was a CPU usage by a runaway Matlab process, but when we killed it there was no effect. We then poked through process lists, thinking maybe we were getting DOSed on SSH connections, until we found three people running sqlite. As soon as we killed the sqlitesw, logins on Linerva came back. We noticed the sqlite processes because we were receiving spew about byte range locks. We thought maybe AFS had implemented these with a spinlock, which was causing severe performance degradation, but this was not the case. We did find out that this was due to a class which was recommending its students access a 700MB sqlite file over AFS. Other ===== Other Other ========== Elections ========= Chair ----- [reading of constitution] kaduk: I nominate adehnert. [dwilson: Second.] adehnert: I accept. kaduk: I nominate dwilson. [adehnert: Second.] dwilson: I decline. adehnert: I nominate bbaren. [zhangc: I second.] bbaren: I accept. adehnert: I nominate ezyang. [zhangc: I second.] ezyang: I decline. dwilson: What would you do differently from your predecessor? adehnert/bbaren: *shrug* adehnert: SIPB seems to be rolling along healthily and unchanged in your tenure. I'm not sure. bbaren: I would tend to agree with that statement. I think David has done a fantastic job as chair. I feel we might want to raise more interest about SIPB; things seem a little bit quiet here right now. jhawk: Can you discuss your experience with respect to SIPB administration, and how this might inform your leadership. adehnert: Previous vice chair. I've largely taken a hands-off support, due to a mixture of various things, partially because things rarely come up that the vice chair does. Hackathons, talking to people, etc. I've held other administrative positions in other groups. bbaren: I have not been in an official SIPB related leadership role, neither as a SIPB EC member or a project lead. However, I have helped run the activities midway for several years, and I also feel like I talk to users a lot, and I feel comfortable about my ability to talk to people who are not members of the SIPB. zhangc: Since neither of you want to improve things, can you list some of the things that David has done particularly well? adehnert: One thing that I can say is that SIPB is somewhat weird, where it is a loose confederacy of various services and members who do their own thing. There's a quite significant role to be played in terms of the individual subareas in terms of project maintainers. But it's a lot less critical that SIPB leadership do specific things. bbaren: I think we're in an interesting time, where we have a lot of people who have been project leaders and primary contributors who are graduated or taken leave over the past year. So I feel like we're in a transitional period as far as project leadership goes. It's been nice to see that this has proceeded smoothly over the course of the year; I hope this keeps up. I would hate to see services have to disappear because of lack of maintainers. dwilson: A lot of duties of the chair are somewhat administrative in nature, and you need to coordinate people to some extent within SIPB, and also with other outside entities. Could you describe experience, feelings or philosophy that you have in coordinating groups, interacting with IS&T, the ASA, the MIT administration, and any other major groups that are relevant? bbaren: I am the SIPB representative to ISTAB. I would like to believe I'm in good graces with jdreed, based on which I feel fairly confident in my ability to interface with IS&T. I also helped organize computer tours, so I have some experience working with the less public elements of the IS department. As far as administration goes, I can't really speak to that in great depth. As far as administration goes, I can't speak to that in any depth, but I feel adehnert: I have less experience working with IS&T than I would prefer; it's something I've done a bad job of doing. But I serve on ISTAB so I know jdreed somewhat. And I also know a certain amount about IS&T. I have a quite strong relationship with ASA. I know a bunch of the people in SAO and CAC and Schedules moderately well, which is useful for more student group components of our operations. Also, says something good for my ability to work with people when I get into that. ezyang: What do you see as the most important role of the chair? adehnert: The chair historically schedules membership elections, keeps an eye on prospectives, sometimes pokes in private, also, I think in the last couple of years, making sure rush happens, and making sure people take on that. I think that sort of rush type stuff is the most important. bbaren: I feel that the chair's general relationship with prospectives and people who may be interested in becoming prospectives is really important. We really should be more aggressively seeking people to get involved. As a chair, it can play an active role, present a nice public face. ezyang: In terms project maintainership, that's something that traditionally is left to the projects themselves, not the chair, what is the influence the chair should have in that sort of context? bbaren: I guess I should have clarified that. I absolutely think the projects should maintain the level of autonomy, but I think the chair should be intimately involved with prospectives and natural members, it should play some kind of role pointing prospectives towards members, and also projects which need maintainers. adehnert: A big part of the role is sort of talking with projects about what they can do to do about recruiting, encouraging them to reach out to people, and on the flip side pointing prospectives at projects which seem like a good match. I think it's important when doing that you don't overstep and point prospectives at projects that really don't want new people working on them, for various reasons, or try to particularly identify individual tickets, or toss prospectives at them to make sure that's a productive thing. jhawk: Absent steps along executing a deep vision, I think one important role of the chair is dealing with horrible crap that comes up. When it doesn't happen, that's great. When it does, it's important. What do you have to say about your crisis management skills? adehnert: I've largely avoided having any particularly exciting crises to deal with. I've done a certain amount of talking with people about administrative changes, through work on the UA, ASA, LGC, which interacts with administrative changes, and has had various things come at them recently. I've also done a certain amount of dealing with things as they come up. But I don't have anything that I'd really describe as a crisis. bbaren: Thankfully, I haven't had to deal with a crisis of extraordinary magnitude yet. I was in the office when Aaron Schwartz's netbook was removed by the campus police. I remember talking with the people in the office, and talking with jhamrick who had still been chair, about how best to proceed from that, so I have some observational experience, but I wasn't the one who made sure that the board made it through being in an unwitting accomplice. kaduk: Followup: suppose you are chair and some crisis of serious proportions has occured. What resources would you find yourself consulting; written documentation, or in terms of human resources? adehnert: Older members are good resources in terms of knowing history of how similar issues have been dealt with; previous chairs, previous involved people. Depending on the type of crisis, talking to contacts in IS&T. If the crisis is like, if we don't want to deal with SMR tomorrow, talking with IS&T and understanding what the actual problem is likely to be fruitful. For other sorts of crises, that's less relevant. Thinking about stories that I've heard before; if I know something happened similar before, using discuss archives of mailing lists, minutes. bbaren: I would be especially willing to consult previous chairs on issues. Also, as you mentioned, talking to jdreed, esp if it's a crisis that involves more than SIPB itself. For things that break, there's who fixes what and the correct procedure for paging people is pretty well established. But I think talking with previous chairs and more senior members, with IS&T if necessary; also, I've found old minutes to be extremely useful. kaduk: What do you see the rest of the EC in advising the chair in certain situations? adehnert: I think it depends on what the crisis is. There are occasionally really sensitive crises. I've certainly got the impression that in some crises they weren't comfortable talking to other members of the group. I haven't run into this, so I don't know if this was a judgment difference. Depending on the vice chair, they would be good to talk to. bbaren: My impression is that actually convening an EC meeting can be somewhat of a hassle. There's certainly a time and a place to do that, but depending on the magnitude of whatever crisis is being dealt with, or whether or not it's something that affects the board as a unit, or really penetrates deep to within specific projects... I think it's a question of is it worth going to the trouble of trying to get an EC meeting together, or if it will be more beneficial to the board to have a decision. kaduk: Sure. It's also possible adehnert: It's rarely the choice I would take to convene a meeting; but I would be more likely to email people, zephyr people, contact the whole EC at once, or contact members of the EC. dwilson: In broad strokes, what are each of your plans for the next year. Are you around over the summer? Are you taking 43 classes? adehnert: I will be here this semester, taking four classes, one of which is 18.100C (15 units) and one of which is 6.UAP (6 units) but probably not a time-consuming six units since it's an MEng thesis proposal. The more interesting aspect is I'm doing 6A, so I'll be in California over the summer and fall, so I would step down at the end of the semester. bbaren: I'm taking four classes (42 units), but one of them is 6.115 which is more than 12 units, but I don't anticipate being a more difficult, or more overwhelming semester than average, and I suspect this semester will be slightly lighter than normal. I don't know if I'll be here over the summer; there's a pretty reasonable chance I'll be here this summer. I will be here for the next fall; I don't plan on taking 43 classes next fall. In interest of full disclosure, I will probably be taking 6.828, but I don't know what else is on my schedule. jhawk: adehnert, do you have a response to the criticism that you may be spread to thin? adehnert: That's reasonable. I think that's less true now than a year ago. I'm doing some things with ASA, I'm doing some things with ESP, but then I'll go totally inactive. I'm currently LGC speaker which is very little time, giving up in a few weeks. And, I'm finishing up being ET treasurer in a week. So it's just ASA and SIPB. kaduk: We've mostly established the chair has an important role to play, in drawing new prospectives in. How much time do you see yourself spending in the office, and being available to interact with prospectives? bbaren: I feel that I spend a pretty good amount of time in the office. I like using mkc because it has a nice keyboard and large monitor. I would say I do most of my work here. I am pretty frequently here on weekends, often at times when there are not other members around. I don't usually stay too late, but this term I'm frequently here in the mornings for a half hour to hour. adehnert: I live off campus, so when I'm on campus and not doing anything my home away from home is the SIPB office. I'm here a fair bit. It's less this semester... well, might be less this semester than last because almost all of my classes are on Tuesday and Thursday so I don't come to campus until really late. I've been here Tuesday/Thursday afternoon for the last week. I'm usually not here on the weekends, and frequently if I'm here I'll be here for an hour to seven to eight. ezyang: Would you be willing to change your behavior, if you thought it was appropriate? adehnert: I would be willing to change my behavior if I thought it was important? But I don't think that current behavior is likely to cause problems. bbaren: I think I'm open to convincing that I spend too little or too much or less than ideal times in the office. adehnert: How interested are you in this position? bbaren: I'm definitely interested. Being SIPB chair is an honor, and honestly not one I anticipated being nominated for. adehnert: Have you looked at the candidate list? I think you're one of the more involved people. bbaren: I would be thrilled to be SIPB chair. I really do enjoy talking with prospectives, and people outside the board, and being chair would be a way to continue doing that in an official capacity. adehnert: I think it could be interesting; I'm not totally sure I want to do it. I think there's a lot of room in terms of how much effort I could put into it; I think doing the minimum would be relatively little reward; putting a lot of effort I could effect change, it would be satisfying, but that would require time and convincing myself to actually do it, and I'm not sure I would be successful at that. dwilson: If elected chair, where would you keep the gavel? adehnert: I do like the dragon. It's pretty classy. zhangc: Is the dragon going to stay? dwilson: The dragon is open to staying. adehnert: If the dragon stayed, it would stay here. There's also the weird hoops. jhawk: Nine-track tape, write-enable rings. bbaren: I think the dragon is pretty classy; I would also be interested in retrofitting the shark-banana to hold the gavel in its mouth, for some definition of mouth. adehnert: Do you have any thoughts about how to make SIPB an easier place for people who come into MIT with little knowledge of computers but lots of interested to get involved with SIPB projects? dwilson: Alex, why don't you answer first? adehnert: I've done a little bit of thinking about this. I talked with a couple of other SIPB members about this, in the fall. I'm not sure what the right way to do it is; I know a couple of years ago we tried a mentor program, where interested SIPB members and interested new members would get paired up, and the idea was that the mentors would help the new members find a project that was interesting, and make progress to it. But that fell apart due to lack of interest from the mentors and failure to follow up. I think that could be an interesting thing to try again; that would require more attention in terms of poking the mentors and making sure that they haven't forgotten. I think making a push to have projects make it clear what good starter tasks are, I've put some effort into doing this with Scripts in the fall. So we have a dozen or so tasks selected that are helpful, and ezyang and achernya and other people helped did. I think a decent amount of the problem is that SIPB is currently not optimally set up to reach out to prospectives in terms of encouraging them to do stuff, except for a few specific times like hackathons, so it requires that if you want to get involved, you have to keep reaching out to older members, poking them, and encouraging them. If you're a quieter person not good at doing that, and it's harder if you need more help, you need to reach out more often. bbaren: I think we've got a really well thought out and well orchestrated formal education problem. We have cluedumps and SIPB IAP, but I think something we could really improve on is our informal or less formal means of getting people involved, or getting people more knowledge. What I saw a good amount of last fall, a good number of people at the first meeting who never been here before, and then the numbers dropped exponentially, either due to hosage, but I'm confident at least one or two cases, there were people here who were genuinely interested in learning about UNIX-like systems, but didn't have a lot of base knowledge; in one case didn't have a lot of programming knowledge. We as a group could have done better to try to give those people knowledge about the projects, and tasks they could do with the projects. So if hey, you want to learn how the system works, I think we could do a lot better job of getting people more general Linux-y computer knowledge. Vice Chair ---------- [reading of constitution] ine: I nominate achernya. [dwilson seconds] achernya: I accept. bbaren: I nominate phurst. [dwilson seconds] phurst: I accept. dwilson: What is your schedule for the next year? phurst: I have 5.5 classes (66 units), one of which is 6.006 which is nothing, one of which is a fairly easy HASS, and one of which is 6.UAT. So I don't think I'll be too hosed. I'm interning at Google in the Cambridge office, so I will be here. achernya: I'm doing 60 units, which is down from previous 78. 3 are course six, the hardest are 6.046. I'm not terribly scared of the other two. I have a HASS class, for which I'm working Dormbase. 4.60. I'm not here over the summer, but I expect to be here next year taking 60 units. kcr: What are you doing over the summer? achernya: I'm going home to Ohio, may be working on some Android hacking. kcr: So a decent network connection. achernya: Good enough to make Scripts Fedora 15 exist. adehnert: What experience do you have with SIPB projects? phurst: Zephyr-IMAP notifier for people who are on exchange. But in terms of Scripts/XVM, I haven't had any experience with those. achernya: I'm current Scripts Architect. Also maintainer of Linerva, trying to coordinate transition to dialup. I have some code in Debathena, but not XVM, but have a significant experience in all the servers. But I'm still frequently corrected by "the actual maintainers." ezyang: What do you see as the most important role of the vice chair? achernya: I think the most important is coordinating hackathons. I think we don't have enough of them, and we definitely should, because that's a primary vehicle by which we get people. I think SMR and projects are less important, since projects tend to be pretty phurst: I'd definitely like to see more hackathons, see more people in the office. I think it would be nice if we had hackathons which weren't in here, because it can be kind of crowded, and I personally haven't come up . j ezyang: Do you think smaller hackathons are a better idea? phurst: I think that's definitely an idea; I think it might be a good idea. If there was a hackathon every more often, then not as many people would show up. achernya: I think the right thing is to have big hackathons, and project specific work sessions, which have been discussed but never implemented, except the Debathena bugsquashing hackathons. bbaren: Also the occasional Gutenbach hackathons. ine: What personal characteristic or experience do you bring which you think uniquely characterizes you? phurst: I used to be a Wikipedia adminstrator. So I think I have a lot of experience dealing with really technically minded people, and the challenges thereof. achernya: I can't top that, I will say that working on Scripts has exposed me to some very different technical perspectives, ranging from "just make it work" to "no you have to do it right." Also, repercussions of certain things that have to be done right. I think there are some places where you can take risks and not think things through (schedule hackathon next week), but other things need to be coordinated and get approval of people, and I'd definitely be willing to do that. dwilson: You are running for vice chair. What is your favorite vice, and what is your favorite chair? phurst: I used to have one of those bouncy balls you could sit on; their very comfortable. Vice, Haskell. ezyang: That's not a vice, that's a sin. achernya: I'm a fan of that *points* chair in the office. I like vice scripts, they're kind of cool. Also, Fedora. Except maybe Mitch. Treasurer --------- [reading of the constitution] ezyang: I nominate kaduk. [bbaren: seconds] kaduk: I decline. ezyang: I nominate adehnert. [bbaren: seconds] adehnert: I accept. dwilson: I nominate ine. [achernya: seconds] ine: I accept. achernya: I nominate lapentab. [dwilson: seconds] lapentab: I decline. dwilson: Plans for next year? ine: I'm taking 72 units, probably will stick with that. So I'm curious as to how much a time commitment treasurer is. kaduk: It's not a large time commitment. ine: I can handle things if they're spaced out and have atomic blocks where you can do the work. kaduk: If you approximate it as two hours every two weeks, that's not terribly inaccurate. ine: Next semester is less so; on the order of 50-55 units. Summer, I won't be here; I might be here on the weekends. I'll be in Rhode Island. dwilson: Do you have any relevant experience you'd like to bring to bear? (Treasurer of another group, or something administrative?) ine: Nothing immediately relevant to treasurer comes to mind... I've worked with Excel. adehnert: Do we have an accounting system we use? kaduk: We leverage MIT's SAP deployment. adehnert: I've been treasurer of [large list groups (5)]. At my peak I was managing just under a million dollars a year, depending on how much you want to do. But I've never dealt with an account that is managed through IS&T, so that will be a fun and exciting change. All groups I've managed have been through banks or SAO. mitchb: I don't actually care, but what are each of your estimates of when I will get reimbursed for the 2 CMOS batteries that I bought three years ago? kaduk: I have the receipts. adehnert: I'm not prepared to make an estimate, without knowing what has been handled yet. [kaduk gets up to get the receipt] kaduk: Basically, the problem is that three dollars and ninety-eight cents. mitchb: Well, this got harder when i started getting all of our disks replaced for free. adehnert: 3.75, is the worst that I've done, so 3.98 is not the worst I've done. Mitch, do you have direct deposit setup? mitchb: It's kind of confusing, and we can discuss kaduk: The accounting question is not ezyang: As treasurer, you will frequently have to navigate the perils of corporate finance and other goodies like SAP. How do you plan on learning the ropes? ine: I would consult with the current treasurer, also my roommate's been treasurer for several student groups. So I would probably bug him if I couldn't figure this out. kaduk: There is in fact some form of written documentation. adehnert: Oh, did you finish it? kaduk: No, I started it... adehnert: I remember hearing about that two years ago... dwilson: Can you describe your motivation for this position? adehnert: Somebody needs to do it; I'm vaguely working to collect all of the student group accounts that I can; I've already done SAO accounts, bank accounts, SAO reserves... this seems like one of the few opportunities to grab an IS&T one. ine: I think it would be a good opportunity to get involved in some SIPB administrivia, although there may be better ways of doing that. adehnert: You should come find a project to work on! ezyang: Wanna join Scripts? kaduk: Here is a question for adehnert: what is the most painful thing you've dealt with in SAP? mitchb: My direct deposit situation. adehnert: One thing that was painful, back in the day it was possible to find MIT ID numbers by going to SAP and searching for people, and one of the interfaces exposed it. mitchb: Only people with SAP bits; if you didn't have SAP bits it wasn't in it. kcr: Bug or feature? adehnert: This was inconvenient because it wasn't conveniently scrapeable. Which was bad because CAC once wanted MIT ID numbers in order to add office access. kaduk: Doesn't CAC still want it? adehnert: It's complicated. zhangc: Could you elaborate on your adminstrative experience? ine: Recruit other people into the Simmons tech committee. That's been remarkably successful. We've coordinated a lot of improvements, we've not touched old things and made new things. adehnert: What year are you? ine: Junior. achernya: We are currently faced with having kaduk been treasurer for the past four years. So it's been a while since we've had a different person; so hopefully the documentation is good, but it also means that the person working on it is the same, so a lot of things have come absolutely normal to them. Do you consider having more or less experience to replace kaduk to be a good thing? ine: From a docu standpoint, less experience is better, if you need to rewrite them. I would schedule time to have kaduk walk me through the top ten things you need to do, write things down, and write that down. adehnert: Shrug. Probably more experience will result in fewer short term screw-ups, but less experience might result in better documentation. To the extent that there is not tons of need for special stuff. ine: Well, I doubt it changes that much. adehnert: I don't expect the documentation to matter too much. achernya: The reason I currently ask is experience seems to be a very big deal. jhawk: I think you've misinterpreted the situation. kaduk: To be clear, when I was first elected, I had very minimal experience, certainly no experience with a system like MIT's. My predecessor did not provide me with very much in the way of documentation or training, and I don't think anything terrible happened. adehnert: I didn't get any formal training when I became ET treasurer. Secretary --------- [reading of the constitution] lapentab declines dwilson declines phurst accepts ezyang accepts adehnert declines kaduk declines dwilson: Schedule for the next year? ezyang: I'm going to grad school, not MIT, and I'm going to PhD open houses on a bunch of Mondays, so I won't be here too much. I may be here over the summer. phurst: Like I said, I'm taking 66 units, but none of them are on Monday nights, and I'm interning at Google Cambridge this summer. ine: How fast can you type? ezyang: Fast. phurst: Pretty fast. I'm not sure specifically. adehnert: Have you taken minutes before? ezyang: Yes. phurst: No. mitchb: You can get trained on this. We trained Donald [Guy] when he became secretary. dwilson: What are you most excited about other than taking minutes? ezyang: Being on the EC. phurst: There are other duties? dwilson: You /actually/ member people, which is kind of neat. ezyang: Oh man, I didn't have to take minutes for this one! adehnert: That's why you run for these things! dwilson: How enthusiastic are you about this? ezyang: Seeing as it took three nominations to get me to accept.... phurst: I'm fine with it. I'm not, like, "Oh my god! I get to be secretary!" But I'd be up for it. I like greeting new people. ezyang: See, you should vote him secretary. Members at large ---------------- [reading of the constitution] Eligible members are: adehnert 8 afarrell 0 cesium 1 dwilson 9 ecprice 0 ezyang 7 iannucci 3 jhamrick 3 kaduk 7 lapentab 6 rye 1 jhawk: Should SIPB have a wiki? adehnert: We have one, it's our website. mitchb: It's called "icky"-wicky adehnert: Also Trac. jhawk: For broad administrative functions, ranging from service to maintainence to gavel holders... bbaren: I prefer to keep things in AFS. lapentab: I think it would a good to call prospectives members, and members keyholders. It sounds better. dwilson: It is true that SIPB uses the term "member" in a very specific way which is different from how many student groups use the term. kcr: It would take a constitutional amendment to fix that. bbaren: I would be in favor of discussing this in greater depth on sipb-office. I feel that there are a large number of people here who have opinions on this, and can attest to previous discussions. dwilson: I think there were efforts in this area that ended up with the interested designation, but it started with vaguely similar concerns. But this has been recognized in the past, and we can continue to discuss it. meeting adjourned at 21:35. Minutes taken and submitted by ezyang.