Summer Schedule, Tech Infrastructure

[rax]

First, summer scheduling: In a change to my schedule, I won't be at Readercon this year. If I'm at any con this summer, it will be Anthrocon (who else is going?), and I"m not sure about that either. Still trying to figure out remaining travel schedule, especially as concerns weddings. When I have a complete calendar I'll probably post it; term-time travel will be limited to random "Surprise, this weekend I'm in X" sort of things planned at the last minute based on not having anything due on Monday. And now, I am going to dork out or a while.

So, tech infrastructure. If you're reading this, you're probably already invested in using computer technology in order to engage with your social network. [0] You probably use a number of different technologies to do this, most of them supplied by socially and geographically distant corporations. These corporations probably range on the evil scale from Facebook's "Privacy is for losers" to Google's "Don't be evil" or Dreamwidth's "We are made of puppies." As much as I rag on Google (and I think they deserve it; a company that large does more evil things in a day than I will do in my entire life, unless I really start trying), they do try very hard to give users a positive experience for engaging with other people on the Internet, and the levels of adoption of their email services, chat services, and other offerings are a testament to that.

That said, I'm sometimes surprised by how many of my friends, Linux dorks in particular, use services like this. A lot of us talk big about peer to peer and community owned infrastructure when it comes to things like BitTorrent or distributed computing, but I haven't seen many projects looking to set up this kind of architecture for things that we use the internet for most frequently, like email, social networking, or blogging. The Diaspora project (distributed Facebook replacement under development) is one counterexample that has gotten a lot of press, but right now it's just an idea. I know a couple of people whose LiveJournals are secretly something else, but for the most part we either just use LJ/DW or have an external blog that shows up as a feed and then a reading account. There are some other LJ-alikes (InsaneJournal, JournalFen, and so on) that may have traction in specific communities, but they're still not quite what I'm thinking of, because...

I really value knowing my service providers personally. Just like I know my bike mechanic by name, drink beers with him, and sometimes just show up in his shop to talk about whatever, I want to have this sort of relationship with the people who provide my email service and other technical infrastructure --- when I'm not just doing it myself. When possible, I think it's awesome to trade these kind of resources either for skillshare or for cost. In some cases, I've been successful with this, or I'm successfully the person who other people come to for this: I co-own a computer in colo with sixolet , and a mutual friend helps us with infrastructure in exchange for backup space, and we lease out virtual machines to our friends at a rate that exceeds bandwidth enough to cover the cost of the machine in, oh... ten years? At the very least it pays for hardware upgrades. [1] This is awesome, and I want to do more things like it.

Some of the things I might want to do are very hard, either because they're just technically very hard (oh my god running a mail server was such a pain last time I tried) or because the protocols are closed (I can't just run my own facebook, because real Facebook won't talk to my facebook, and so I can't get messages from all my extended family who refuse to use anything except Facebook to talk to me). But some of them shouldn't be that hard, and might be of interest to other people, and I wanted to write about a couple I'm hoping to do and get comments and suggestions on them:

• Cohousing wiki-type infrastructure. I imagine this as great for everything from grocery lists and chore structures to shared projects like "Let's all have an awesome event that requires coordination!" and want to set it up for my new house. I know a bunch of the random warehousey things around do this --- Langton Labs, for example --- and I think some smaller apartments (Technodrome, right?) do this too. I heard recently from a friend that she and her partner used Jira to coordinate just between the two of them. [2] So this is clearly doable --- but I don't know of any best practices anywhere, or templates, or anything like that. If you do this, what works for you? What doesn't? If you'd like to use this but don't now, what would encourage you to start? Would you want a template? Do you already have a server to run it on?
• Mailing lists. The commercial-free services like yahoo groups are freaking abominable. Most organizations seem to run this by setting up mailman lists --- I tried to set up mailman and gave up after around ten hours, although this was a couple of years ago and maybe I should try again. (I still get a bounce message in my inbox from that mailman install every day. It's not worth the effort to figure out why.) Most of the social groups I know either do this client-side (some email clients kindly track lists for you) or through the MIT mailing list system. Since I'm now two universities, six years, and a thousand miles removed from MIT, I feel like I should be running my mailing lists through something different. Is mailman the state of the art? Are there other tools I should be looking at? Are there people out there with semi-open mailing list services, or people who would use one if it existed?
• Event invitations. I traditionally do this via mailing list, but I've identified two big problems with this. First, for events that require RSVP/guestlist, technology could help a lot with tracking this --- and Evite and Facebook handle this sort of thing in a way that people understand and are arguably coming to expect. Second, I increasingly have friends --- people I'm quite fond of and want to see --- whose email addresses I don't have, and this causes me to miss them when I send out party invitations. (Hi guilwolfie!) I know at least one person has rolled this on their own, because I've been invited to a party that used it, but I don't think it was open source or know the author to write and ask if it's something other people can use. Also, it only worked over email. :) I think it's important that a tool for this contact users where they are, whether it be AIM or Facebook or email or whatever, and not require a new account. I don't really know how to do it, but I know that I want it, and I'd love to hear other people's thoughts on it.

I would really love to see community owned and operated email services, too, but I think that's very hard --- gmail does a better job both in services and in interface than I think I can do, and I don't think "run by your friend and not by Google" is enough to overcome that with anyone except people who already aren't using gmail. The three above, especially the first two (if your social group all uses facebook, the third is basically solved) I think are particularly worth looking into because I think we can build something that is better than the current alternatives, not just more open, and I think it may not even be very hard. Does anyone know if SIPB is working on any of this kind of stuff? It seems right up their alley...


[0] You might also be a search engine.

[1] I don't want this post to get mega-technical but if you ever want suggestions on setting up something like this, let me know. It's definitely doable; there are more people who would rent virtual machines if we wanted to rent more.

[2] I've poked at using RT for this personally but it was too heavyweight; I've poked at Hiveminder but it was too lightweight; I've considered using Salesforce case tracking but the version I like that I use at work is $999/year. I'm still using flat text files, and this makes coordinating with those close to me difficult sometimes.

Comments

[rivenwanderer]

Yeah, I have mixed feelings about software-as-a-service. I really like that I didn't have to shave hardly any yaks to get my house's online organization set up (using Google Groups, Docs, Sites, Calendar, etc), but there's no good reason that these things need to be hosted by someone else. I think that hosted VMs for everyone would be really cool and could lead to a lot of neat stuff that also came with better privacy management... I wonder what it would take to get to the point where running your own hosted VM off in the cloud somewhere was no longer a sysadmin-pointy-hat sort of thing, and was considered no stranger than having a personal computer.

[keshwyn]

The fact that I really don't trust Google is a prime reason I'm currently working on getting a Ubuntu VM up at home, so that I can install Zimbra. It does IMAP/Calendar/Stuff for me, and if I can get my friends on it, we can all share our calendars and know exactly who has access to them...

...but it's not big enough for the world.

Mailing list software is really, really not for the faint of heart. Personally, I use sympa, but that's mostly because it'll do LDAP-auth, which I always think is a good thing. "Here, let me get your password set up such that you don't have to remember each and every password for each and every list...it's just the same one you use over at [thingy]." If I could figure out how to tie it to openID, so much the better.

But sympa's even more heavyweight than mailman - it requires a mysql backend.

[rax]

I don't actually want big enough for the world; my sights are not that high. :) Zimbra's what we use at work, and I'm relatively fond of it, I may have to try setting it up on one of my VMs. Is there sharing between instances, ACLwise, do you know?

An OpenID-backed mailing list software would be made of kittens. I'm not sure I have the chops to write it. I wonder if it's something I could put a bounty on?

[picklish]

I'm curious about what you're getting out of knowing your service providers directly. In particular, knowing your bike mechanic seems more like knowing the person writing the software that you're using, and less like knowing the person handling your hosting and infrastructure. Just because I'm hosting your mailing lists and your wiki and the source control for your recipes doesn't mean you can ask me for any special features you need or that I'll know enough to support your edge cases or even that I'll have time given that it is almost certainly not even a part-part-time job. Is it more about the joy of skill-share and bartering? Or, just a desire to strengthen friendships through (essentially) business relationships? I guess I'd just be interested in hearing more about the particulars of how this appeals to you.

Also, despite doing the backstroke in a pool of Google Kool-aid, I do believe the right approach here is to have more decentralized services, like email and XMPP. In the same way that I don't want to join the Facebook, there are plenty of folks like you who would rather not use Google products. It's just unfortunate that there's no open protocol for data sharing across servers for these various services that you list. Having to convince somebody to sign up on some random website on top of just having to convince somebody to use a service is usually far too much of a burden. Neither Google and the Facebook's solution of just making their walled garden larger nor your solution of just setting everything up locally really seems like a great answer.

[rax]

Well, my bike mechanic didn't build my frame or my handlebars. (He did build my wheels, but that's not normal.) If you're hosting the source control for my recipes, I'll already be talking to you on a regular basis, and so I can ask "Hey, is it easy to implement X thing?" and you can either say "Sure!" or "No!" or "Ask me again in couple of months when I'm done moving!" and all of those answers are cool. If Google's doing something in a way I don't like, I'm utterly fucked, I won't even get a response, they may change it at any time. My friends won't do that. And of course I think skill-share and bartering are awesome, and I think friendships are strengthened through having mutual minor obligations and favors. (This may have come from the environment I grew up in.) I like trusting the people I have business and infrastructure relationships with; I'm comfortable growing into that trust with a service provider, but something like Google or Facebook is just too big.

I'm 100% with you on open protocols and data sharing --- I'd much prefer things like email and XMPP. But I think some tasks are best set up as local things (like a house wiki) and others just aren't going to be distributed in the near future, and I still want to perform those tasks in a way I feel comfortable with.

[shatterstripes.livejournal.com]

In my experience wikis really don't work for scheduling, etc types of things, because they don't reach out with email and say "HEY I GOT MODIFIED COME LOOK!".

[rivenwanderer.livejournal.com]

Yeah, our house wiki is really only used for taking minutes and writing down other stuff that's being talked about in person at the same time. (And it works pretty well for that.)

[rax.livejournal.com]

It's possible to make them email you when something's changed --- our work wiki instance automatically emails you the diffs anytime a page you want to see changes on is modified. I find it really useful for things that change infrequently, but annoying for things I would have been checking every day anyway. I think reaching out in general is valuable, though I question whether everyone would want email; some people might prefer a more temporary message?

disjointed thoughts

[wrane.livejournal.com]

*oh god JIRA
*Technodrome has a PB wiki and in theory it was going to have a grocery list and other sorts of things but in reality, we only use it to click on the link to refresh the server to put more media up. we have a mailing list for parties and occasionally send email to each other but it's hard to make use of a wiki when all four of you hang out together almost every night - so any technology would kind of be a solution looking for a problem. Oh and Chris wrote some financial tracking software that you can use from the wiki. and if somebody goes to the store we ask the living room "hey, anybody need anything from the store?"
*i kinda like hiveminder but never used it once i set it up - i did get emails reminding me to call my mother for like three years though. i know someone who really liked remember the milk, but i can't speak to it. i have an abiding love of chore wars, but nobody loves it as much as i do so i mostly run it just for myself as a motivational tool - someday i will be a Level 20 Paladin of Vacuuming :)
*"utility vs. privacy is hard, let's go make-facebook-accounts-under-not-our-real-names-in-order-to-get-party-invites-and-then-worry-all-the-time-about-our-parents-finding-our-facebook-account!"

[zdenka]

I'm a search engine!






Okay, I'm not really. But I find your use of footnotes entertaining.

[davidglasser.livejournal.com]

Yeah, we technically have a wiki (Wikimedia) at Langton but we barely use it. Unfortunately the house is small enough that people feel like we don't need too much formal organization and record keeping but large enough that we really do.

[sprrwhwk.livejournal.com]

I've found running a mail server on Debian to be relatively straightforward, with the caveat that I outsource my spam filtering to Lux Scientiae, Inc. (https://luxsci.com/) (which is apparently one Erik Kangas (MIT PhD), and Somerville-local) for $10/mo., because maintaining my own SpamAssassin instance is Way Too Much Work kthx. LuxSci is in its turn a reseller for McAfee nee MXLogic's spam-filtering service, but I've been quite happy with the quality of the filtering (very few false positives) and the service overall. I point my MX records for my domains at their servers and they forward ham to my mail server. For a client I use Sup (http://sup.rubyforge.org), which is basically GMail reimplemented in ncurses, with which I'm quite happy. It has all the fast full-text search and threading and labels and so on you've come to expect from GMail. I've also found hosting my own Jabber server on Debian using ejabberd quite effective and straightforward, and that lets me talk to people on any of the big Jabber services (LJ, GTalk, Facebook, MIT, jabber.org, etc.) or their own Jabber servers transparently; federation is awesome. I use BarnOwl as my Jabber client, so Jabber messages flow together nicely with my zephyrs and twitters and so on.

My house (http://asec.aperturehosting.net) uses a MediaWiki install hosted out of a SIPB AFS locker via Scripts (http://scripts.mit.edu), which works since we've all so far been MIT students and alums. We mostly use it for static-ish data, though, not for coordination -- landlord's contact info, how to set up the printer, that kind of thing. Coordination is mostly over e-mail via MIT's mailing list infrastructure and zephyr/AIM/Jabber. (One of my housemates runs a Zephyr<->AIM gateway which we have set up on the house zephyr class so even the non-zephyr users among us have access to that channel.)

Projects I'm aware of in these spaces:

• It's not specifically a SIPB project, but a few houses of SIPB cruft have put together Bluechips (http://github.com/ebroder/bluechips), which is a ledger setup for cohouses.



• A friend has had good luck with Enemiesof Carlotta (http://liw.iki.fi/liw/eoc/index.html) as a non-MIT mailing list administration tool, though apparently it's not currently being maintained.



• SIPB has been working on an open-source Doodle-killer they call Clockworks (http://sipb.mit.edu/projects/clockworks/), though I don't know what its current status is. I've not yet seen it in use anywhere.



• bluedaisy and nakor use a DaVite (http://marginalhacks.com/Hacks/DaVite/) install for managing RSVPs, though it's also not currently maintained.

For my personal organization I use a combination of my mail client, Hiveminder for to-do lists, and Google Calendar for calendaring. Knowing obra, Hiveminder is a friend-run service, and I haven't yet run into anything that compelled me to switch off Google Calendar, though I think there are some similar host-your-own solutions out there. (GCal integrates nicely with my Android phone, and mostly I just haven't cared enough yet -- I'm just now moving off paper. :-) Except for e-mail, I don't really use those services to interact with anyone, though it's possible with both. In particular, sharing one's Google calendars seems like a boon to scheduling with other people.

Out of curiosity, what do you like about Salesforce that you don't find elsewhere/what did you find missing from Hiveminder? How are you using flat text files currently?

[rax.livejournal.com]

Thank you for all of these links! That's awesome.

Salesforce allows me to track and sort on all sorts of metadata, like "who am I doing this for" and "when did I start" and "what sort of task is it" and "what is the phone number of the person who I am doing this for." Hiveminder does some of this, but not in the same way; part of the reason I'm so tempted by a Salesforce instance is that my brain is already Salesforce-shaped from using it 40+ hours a week for work.

My flat text files have a bunch of different categories at the top (writing, infrastructure, correspondence, random, work, crawl) and then near the bottom a list of day tasks:

MON:
- work
- cancel cathedral utilities as of 6/1
- bring laptop in for repair
- work meeting emails
- morning meeting with $PERSON

TUE:
- work

WED:
- work
- 9 AM $CUSTOMER meeting

The tasks in the upper section get time estimates and occasionally status messages, like:

- BLOCKED ON COLO pay for colo (30min)

And sometimes have subtasks in categories:

- house wiki:
- wishlist
- necessary repairs list
- calendar of visitors
- list of birds seen
- pet care stuff


This works great for me (and is based on what I used to do on graph paper), but while I think it's awesome that people are all "We all live together and don't need infrastructure," I think just eredien and I would benefit from a little more infrastructure here, let alone adding more people to the mix. :)

Google Calendar is awesome but fails my "don't give more data to large corporations" test. If Zimbra talked to Google Calendar so that other people could see my schedule if I ACLed them to it without having to use my Zimbra instance... which it looks like people have done (http://www.unicon.net/node/1300) although hackishly... Hmm. Maybe I do want a zimbra instance...

[rax.livejournal.com]

Google Calendar is awesome but fails my "don't give more data to large corporations" test.

To be clear, I realize that if I make the data available, it's not just mine anymore; but I want to host it or know who's hosting it whenever possible, and I'm trying to pick off low-hanging fruit.

[kelkyag]

I've babysat mailservers before and don't recall it being all that difficult, but it's been a while (pre-mailman). I'd be willing to beat on this again; I would probably ask siderea what she uses for more current advice.

avacon uses something that looks vaguely like evite that's hosted on his own systems for invitations & RSVP tracking. I suspect it's the same DaVite that sprrwhwk mentioned.

On a related "I want this; is it already out there?" note, I've been halfheartedly looking for a wiki that will let me easily make a copy (read only is fine) to throw on a single machine and use locally only, preferably without having to set up a webserver on the single machine. (This is for Madrigal, where I both hate the google sites thing we're currently using, and would like to have this available read-only on site where there is no net.)

[sprrwhwk.livejournal.com]

ikiwiki (http://ikiwiki.info) is kind of that -- it's is a wiki backed by Your Favorite Version Control System, so it's easy to keep a working copy where you need it, net or no net. Doesn't necessarily get you something you can browse locally with a web browser, but it may be enough for your purposes. I've been using it to run my main web site, actually. Being able to publish a blog post by committing and 'git push'ing is really awesome. :-)

[baniszew.livejournal.com]

I would love a non-mit mailing list tool, especially since half of our house doesn't have real MIT accounts and can't update our party list themselves. The main list alternative I've heard posited is google groups, but no one seems to like it that much. I haven't looked at it too much myself, other than having used it for our mystery hunt team. (But in general I am kind of hesitant to turn my life over to google.)

[kelkyag]

Yahoo groups is also not what I want from a mailing list / mail archive server.

[nathanjw]

I probably appear as Mr. Google Kool-aid at the moment, but the things you're talking about have some common properties - doing them as distributed and/or trusted things is hard, in theory and practice both. You have to get all the people you care about to buy into something. Email used to be at least a lowest common denominator of communication, but even that doesn't quite work any more, and any richer experiences require software, somewhere, and today, the way to do that with people on N platforms (including mobile devices) is web apps, which means a server, which means giving up control.

Maybe the future is RSS++/PubSubHubBub/whatever readers for everyone, on their local co-op vegan ISP, but that's not where the money is, so it'll be an uphill battle.

(This low-coherence rant brought to you by homemade ethanol, which is legal, fun, old-school, and a great source of local community. Good beer doesn't travel well.)

[rax.livejournal.com]

I don't think mailing lists require anyone to buy into anything. It's definitely more of a problem for the other technologies. That said, I'm comfortable with web apps on a server; I just want them on my server, or my friend's server, not on Google's server. I recognize this isn't a solution for everyone, but it's what I want to do right now. :)

[iankeith.livejournal.com]

I'm going to AC this year. It'd be cool t ohang out again.

[rax.livejournal.com]

It would!

Now that I'm a little more local, are there other midwest cons I should be considering going to? I know there's one in Indy, but it conflicts for me this year.

[iankeith.livejournal.com]

There's one in the Chicago area in November that I hit every year. Also, Michigan in April is another one I tag.

[cshiley.livejournal.com]

A guy at work put together a suite of best-of-breed, mostly open-source, stuff, and the mailing list solution was mailman.

Service Providers

[krdbuni.livejournal.com]

Prismatic (http://prismaticmedia.com) started life as a way to convince people to migrate off of LiveJournal and over to RSS and such, but people in my social circle seem to have integrated LJ into their lives to the point that saying "I'm leaving LJ" prompts responses of "g'bye" rather than "where are you going?" It's one of the reasons I largely quit posting after setting up the Ranch on Mars. I hated having to manually update the LJ to point to the new site, didn't feel like paying LJ to syndicate the public feed, have too many private filters requiring people have accounts, and have largely come to the conclusion that the emotional price of being evil-by-proxy isn't high enough on its own to motivate most people to stop. See Wal-Mart and McDonald's for larger examples.

Re: Service Providers

[rax.livejournal.com]

I think for me part of the difference with Wal-Mart and McDonald's versus LiveJournal is that LiveJournal has unique content created by people I care about, while Wal-Mart and McDonald's do not. I'd happily pay more to get goods from somewhere less evil, and would happily pay more to get content from somewhere less evil, but this is where the unique content I want to consume is. (I also don't find LJ as evil as other people do, but they are clearly not made of angel-ponies.) I expect this is the case for most people --- although in some cases, McDonald's may have the unique content they want, or the effort investment of not going to Wal-Mart may be much higher due to geographic or financial constraints.

Dreamwidth sort of helps with this, but has a bunch of similar problems; the solution I've seen and liked most made an actual LJ post and authenticated against openID, but when you clicked on the cut tag, magically it wasn't actually a cut tag, it was a link to the off-site page --- sort of like what Dreamwidth does, except homebrew. If it were a higher priority for me (I want to get away from using MIT for mailing lists and set up wiki stuff first), I would probably do something similar.

Part of my problem when people change things is that I find infrastructure change emotionally expensive; when someone says "I'm leaving LJ" and I want to follow them, it can be very hard for me to make following them a part of my routine, which is actually very much set in stone. I've managed to get Dreamwidth mostly into it, but it took a few months of hemming and hawing and researching and then around a month of actually having it set up. It may be that other people aren't like this, but I know I'm very much helped by transition periods between technologies and the option to use legacy stuff for a long time. (I still use twm, and I finally learned how an RSS feed worked this year because I had to for work. Seriously, I'm bad at this.)

Maybe I misunderstand, but looking around in it, Prismatic (http://prismaticmedia.com) seems way more like a webhosting service than like a blog? The things I see hosted there I wouldn't have wanted to put on LJ anyway (though maybe that's erroneous thinking on my part --- I tend not to put creative work here). Although now I know where all your content is! I should check out your fiction sometime.

Also if you want LJ syndicated feeds of anything, I can make them.

Re: Service Providers

[krinndnz.livejournal.com]

The reason that there's still a lot of LJ stickiness is pretty much exactly the same dynamic as Facebook, despite that both are headed deeper and deeper into the service of Mordor. Network effect is big, and so is the fact that it does, in fact, make easy things that are somewhat difficult. You and Jessie have very ably demonstrated that you can make a WordPress blog interact with OpenID in a pretty reasonable way, and replicate some of the permission infrastructure of LJ. It's still a pain in the ass, quadrupled if you're not already a technical sort of person, and it requires going and learning another infrastructure, which as rax notes is a non-trivial mental-emotional burden.

[darthbitsy.livejournal.com]

Re: google, this didn't get much press: http://arstechnica.com/web/news/2010/04/googles-opt-in-dystopia.ars

FWIW, Dispora, in its press, has been comparing itself to wordpress.

[krinndnz.livejournal.com]

This is nifty information that I am noting for my own future reference.

[lhexa.livejournal.com]

Some disconnected thoughts...

• I'm quite in agreement with the thought of people weaning themselves away from the more harmful social networking sites, but when I think about my own experience, their reluctance to do so seems more understandable. When I started using Livejournal, I had no idea that it would become important to me, so my dependence on it increased gradually; similarly, Linux geeks and other open source enthusiasts appreciated right away the importance of operating systems and other core applications, but did not anticipate the importance of social online sites.
  
• That being said, the open source movement created Wikipedia -- it wasn't blind-sighted in that endeavor, like it was for social sites. I don't know what accounts for the difference.
  
• Sure, there's value in knowing your service providers, but to make that a systematic criterion is unrealistic in modern infrastructure. You know your bike mechanic, but do you know the people who built your house, who check your electric meter, who deliver your mail, who maintain your streets, who filter your water, who...? It is a fortunate person who knows even a few such vital people. Right now IT is new enough to have all sorts of enthusiasts, and to reasonable expect to be able to draw on them in need, but I think in a few generations it'll be like car work is now -- sure, there will still be plenty of enthusiasts, but most routine IT work will have been institutionalized.
  
• I know what you mean about the emotional attachment to a service... thankfully, I also have a strong countervaling perversity, which tells me that if there's anything that most or all of most friends not only do, but find almost unthinkable to change, then there must be something wrong with it. Accurate or not, it made ceasing to post to Livejournal easier.
  
• It seems like idealistic types like Paul and me are those most inclined to pull up roots and move when a site becomes objectionable. I say this not to insult those who don't move (the "remarkable apathy" I mentioned in my exiting post, which keeps people from changing the circumstances of their lives, has its reasons), but because it just occurred to me that idealism might be a type of intellectual and emotional nomadism.
  
• Some day I'll be able to leave Gmail, its mildly annoying ads, and its disturbing hegemony behind forever. I'm likely to end up in academia, so I can rely on a university for many of my IT needs. The problem is, I need both a work and a personal account (not so much for privacy as for organization), and it makes more sense to use my graduate student account for work.
  
• Event invitations, mailing lists? Heh, I don't envy you -- the dominant reason why I go to large social gatherings is because that's the type my friends prefer to set up, and thus the only way I get to see many people. With the ability to follow my inclinations, my social life would be a string of small, intimate conversations. I am impressed with how many different things you've tried when it comes to pertinent technology, though.