Today Nat announced this new calendar server project called Hula, and I've got a funny story about that.

Nat was in town, and he stopped by to say hi and chat, and he said, "So we've got this big pile of code we're going to release, and we're going to build an open source groupware system! It's going to be awesome!"

And I said, "Jesus Mother of Fuck, what are you thinking! Do not strap the 'Groupware' albatross around your neck! That's what killed Netscape, are you insane?" He looked at me like I'd just kicked his puppy.

( --More--(12%) )

Tags: , ,

118 Responses:

  1. whumpdotcom says:

    Your friend should speak to <lj user="src">, who has been thinking about the same thing with respect to a non-spammy Evite over SMTP and POP.

  2. usufructer says:

    The "Oh fuck. He's right" part is funny, when viewed from a position of 'not my project'. Which is most people.

    Quality essay. I just re-read all of gruntle, thanks for something new.

  3. i would argue that a lot of reasonable people working in large(r) companies do want the sort of thing it sounds like he's working on. i've been listening to a lot of grumblings lately about how although nobody really likes MS exchange, it does a lot of things in one place that really SHOULD be done in one place that there isn't a good alternative for. i am obviously not a managing asshole type, but i'd really like to see a good "free groupware system." it could make stupid problems at work go away, which would mean i could get home sooner, which would give me more time to get back to the more serious business of getting laid.

    • bdu says:

      i am obviously not a managing asshole type, but i'd really like to see a good "free groupware system."

      and that's a perfect example of why you're not the managing asshole-type. Managing asshole-types BUY THINGS, and the more ubiquitously used this THING is, the more likely they are to sign on the dotted line. They are inherently distrustful of valuable products that cost them nothing and are an alternative to the big package that everyone else plops down $300 a seat for. IMHO this is the crux of why any such project is pointless.

      • Managing asshole-types BUY THINGS,

        I wish my manager was an asshole so he'd invest in a new NAS :(

        • bdu says:

          Let me rephrase:

          Managing asshole-types prefer to BUY software rather than use open source, they rarely spend that money on anything useful like a NAS.

      • lohphat says:

        The reason I have to buy groupware is for the lowest-common-denominator sales and marketing fucktard who can't function without it.

        "Why aren't we using Exchange and Outlook?"

        Because you're too stupid to learn something simpler that doesn't cost us $150,000 a year in licenses.

        "But look! I can sync my contacts to my PDA and Bluetooth[tm] enabled dildophone! By the way, do we have any of those newer, lighter, faster, more fragile, expensive laptops in yet? I have to compensate for my shrunken member playing solitaire while starting at the other guys equipment in business class."

        Group calendaring and workflow are a must because they are useful to some extent, but the real reason is that these cretins are too hard to retrain -- they already carry a quota and use that clout to dictate infrastructure. Usually blowing through wads of cash to do so.

        These people need to be shot.

  4. ged says:

    Novell's HQ is now in no flyover state but... Massachusetts, bluest of the Blue States (that's blue as in "sad", "the state of denial", etc.).

  5. jarodrussell says:

    Thank you. Lots of good points to ponder for me to ponder as I try to streamline the homegrown E-Learning and intranet system at the school where I work.

  6. violentbloom says:

    But we love <lj user="wscullin"> anyway.

  7. harryh says:

    Do you ever miss working on this sort of thing full time?

  8. ammonoid says:

    Ok, I feel like a tard because I have no fucking clue was groupware is.

    Also, I drive up 101 every now and then and get to wonder what the fuck "middleware" is.

    In my universe "middleware" would be a waistcoat or something.

    • stlst says:

      From the sound of it, "groupware" is one of those hazy names that implies it's a social-sorta thing. Not really sure, though...

    • Middleware is the thing between you and something else that is supposed to make it easier to use, and sometimes does. Just as often its the opposite.

      Really you could consider the livejournal protocol stuff middleware. It allows people to write clients to an interface without worrying about what server is on the other end (e.g. deadjournal, etc.), and vice-versa. Not a very complex one, though.

    • jwz says:

      "Groupware" means "one giant monolithic, proprietary piece of software that is your mail server, calendar server, conference room scheduler, file server, task list, bug database, and kitchen sink."

      "Middleware" means "I am locked into a closed, proprietary application from vendor A, and a closed, proprietary application from vendor B, and they don't interoperate at all. So I buy something from vendor C that kinda-sorta lets A and B talk to each other."

      • ronebofh says:

        "... except vendor A only speaks German and vendor B hates Brussel sprouts and they will only let me talk to them if i'm wearing a blindfold and a tutu, and i'm often recommended by consultants who will charge an arm and a leg and then vanish as soon as i've been purchased and, hey, why is there a label that says 'Snake Oil' on my ass?"

        • "...hey, why is there a label that says 'Snake Oil' on my ass?"

          Wow, you're lucky... apparently you got the 'Snake Oil lube' option, so when their differing formats fuck you up the ass it burns less. I didn't get that option...

  9. Dude. It is possible to get some laid by creating a really really really really really really useful software product. Take single sign-on for example.

    It takes many man hours to configure a server to do single sign-on just right, and by "just right" I mean "linked into AD and Exchange". If there were an open-source project that bridged 20 million different gaps and let you do something so incredibly weird as be productive, for instance in a way that Microsoft paid-for products could not allow you to be, you could immediately switch ignorant users to a new platform just for the fact that they could work easier. And that my friend will get you laid at any geek convention.

    Unfortunately, single sign-on is a myth. You simply can't tie together every competing product in the market and have a true seamlessly working environment, or there wouldn't be any competition anymore really ("who gives a shit about Product B, Product A will interoperate if someone else is using Product B so might as well go with the one that works [A]."). What's my point? I wish I knew... I guess it's that you don't need to only code for the ignorant masses in order to get praise, much less like what you're doing. If it's new and innovative (to the person) they'll hack away at it as long as it takes.

    Asterisk is the perfect example. Did that fucker Mark have hundreds of girls humping his leg when he was developing Asterisk? Nope. Were there thousands of soccer moms begging to get ahold of his src? Nope. However, if I didn't suck ass at C I would have tried to help him out with development when the project wasn't quite popular yet. Even when I didn't have an FXO card I thought it was a cool idea and wanted to try it out. Now he's the king of Voipland, and dozens of hackers and coders want to use his product.

    Now, it is difficult to encourage people to work on Groupware suites, mostly because you already know it's a very large task and nobody wants to work on a large task alone. That doesn't mean you couldn't go about it in the right way as to entice 22 year old virgins to get the ground under the project's feet. For instance, if you could develop an app that was not only as intuitive and powerful as Outlook but was prettier and more fun to use and you knew you could switch every single moronic manager to it just by looking at it, that would be one heck of a project to be a developer on. It's like the distros; if there's a sweetass idea people will flock to it in droves. Gentoo has probably accumulated developers faster than any other distro that has lived. (they also probably have much less stringent requirements on their developers, as you pretty much only need to know bash and a little python to modify something, but you get the point)

    • valentwine says:

      And that my friend will get you laid at any geek convention.

      Leave your fucking house. Right now. How long has it been?

      • hepkitten says:

        Also hello ConWhores are the WORST. Who wants to ride the town bicycle?! Especially after watching it get ridden by three other fat losers the day before!

      • Are you saying if I created an app that was simpler and at the same time more powerful than any Microsoft product in the world you wouldn't wanna tap this pale white ass? Come on now. Stop frontin'.

        • No, we'd want to tap the sexier, more businessy savvy, Steve Jobs type that marketed it for you and made all the real money. Only nerds want to tap The Woz.

    • pberry says:

      I doubt CAS from Yale has gotten any of those people laid...even though it can be beaten into PeopleSoft. But who cares if it was single sign-on, you're still in PeopelSoft and thus you aren't going to be doing anything "productive".

    • lherrera says:

      It's not about you getting laid, man.

  10. ellyjonez says:


    there is no good group open source group calendaring thing right now. at ye olde EFF we use a real life wallhanging calendar to schedule conference rooms, because there's no awesome open source solution for us. i would leap for joy if nat et al could create something i'd want to use for scheduling meetings n' crap like that.

    OSAF is apparently working on something which might help though:

    • ckd says:

      Back when I worked for EFF, we only had one conference room, so it was pretty easy to handle scheduling it. Heh. That was when EFF was still in Cambridge, though.

    • otterley says:

      Sadly, Chandler has also been vaporware for two years.

      • ixjonez says:

        perhaps it will interest you to know that i've started working on the chandler server as of jan 30. so even if the chandler app isn't progressing as quickly as you'd like, there will be an(other?) open source calendar and sharing server available to help people around the world get laid within the next six months.

        i think jwz is on the money with this post, btw, and from now "how will this get somebody laid" will be my guiding principle.

        • sunsetdriver says:

          they already made ical and the related standards. it's just a file that you can access in the thousands of ways you access files: http, dav, ftp, ssh, nfs, samba, blah, blah-de-blah. and that means the server is already there: apache, ftpd, samba, nfsd, whatever.

          what we don't have is clients. take mozilla sunbird. it supports http/webdav access but it doesn't support PROXIES! i haven't tried the other calendar stuff from mozilla, but it supposedly needs some polishing.

          why, why, why, why, why can't you help them instead of creating yet another half-assed implementation of a caledaring system?


          • ixjonez says:

            perhaps i didn't explain fully enough. the chandler server has two main functions: 1) to provide a caldav server that is interoperable with your calendar message of choice, and 2) to enable sophisticated server features for the chandler client app. only the people who use the chandler client will care about #2, but everybody in the world will be able to benefit from #1. you will be able to download and run a chandler server to give yourself secure calendaring and file sharing with the same effort as (probably less than) installing apache.

            sunbird and ical both support publishing a calendar to a webdav server as a monolithic icalendar (.ics) file. cool. i find that useful. what they don't do is allow you to specify read/write permissions on a per-user basis to each individual item within that calendar, for any user in the world, whether the server knows about them or not. hell, they don't do read/write sharing at all. they don't have any way for you to share your freebusy time with other people in the world, or to use that freebusy time to ease scheduling. they don't have any way for you to automatically figure out where in the world your buddy's calendar is when all you know is his email address.

            fully implementing caldav (and caldav scheduling, when it happens) opens up a new world of calendaring and scheduling that most people haven't even thought about yet. and it enables the implementation of services like an evite clone that can interoperate with other calendaring and scheduling services. neither the clients nor the servers are ready yet (tho some are getting closer). we have to have both.

            (by the way, caldav is a lot more than "http, dav, ftp, ssh, nfs, samba, blah, blah-de-blah". if you haven't already, i encourage you to read the spec and find out how.)

            and hey, i don't do client software. do you? how are you helping?

            • ggerrietts says:

              You know it's funny, about two years ago I started working on Chandler, and about a year and a half ago I went in and interviewed. They were still working on the server, trying to nail down data model, etcetera.

              I walked away from the interview (and subsequently the process) with the impression that it would be years before it was useful because everybody on board there wanted to do old-school software development, not open-source style development. People were too busy having meetings to evaluate or apply the patches I was churning out. Has that changed? Is anyone concerned about the fact that there's nothing useful or usable after 2 years and many hundreds of thousands of dollars?

              That's the bit that I find suspect about Chandler: it's an open source project without open source agility.

              • ixjonez says:

                shrug. i've been at osaf for two weeks. i'm the only server developer. there were zero lines of server code when i got here (so i suspect you are mistaken when you say they were working on it a year and a half ago, but maybe they had something once they threw away).

                i'm about two weeks away from a 0.1 server release that will be a fully-functional multiuser sharing server with administrative web interface (and so forth and so on). next step: caldav. i like to think that i move quickly and push projects forward. so hopefully you won't have any complaints about the server project in that regard.

                i can't speak to the client team. they've made some decisions i questioned, but i'm not a client developer. they know that work better than i do, and from my perspective, desktop application development has always seemed to move at a glacial pace.

                • mcg says:

                  Wasn't Chandler supposed to do all it's magic without a server, p2p style? I kinda always thought they'd never pull that part of the project off.

  11. stlst says:

    What's all this nonsense about designing software for the user? Don't you know there's no money in that? You gotta aim for the Corporations and the Managers with Big Budgets and Little Brains. I'd think your time at Netscape would've taught you something about that...

  12. spoonyfork says:

    You are spot on. If someone could create a open protocol web-enabled version of Microsoft Outlook's enterprise calendaring I would be one giant step closer to nirvana. I have my hopes on Mozilla's Sunbird but right now they're just hopes. Sunbird's calendar sharing is possible but not trivially easy.

    P.S. Ever consider touring the country speaking at universities or large (Fortune 100) companies (paid or otherwise)?

  13. grahams says:

    You are right... I thought "everyone" had acknowledged how lame ALL calendaring software is, and how especially lame non-Exchange/Outlook calendaring software is.

    The trick is, I bet the actual useful "spirit" of groupware will arise from a simple, good, useful app like they are (now) describing with Hula. If people are able to easily publish their schedule and organize events, and this is done with a good focus on the user and usability, then this will be a great way to organize a team as a side effect. And by taking that approach, you toss out all the silty garbage that usually gets tossed into groupware.

    I'm glad you talked some sense into him/them...

  14. lars_larsen says:

    Nat spends far too much time talking to middle managers.

  15. jmhodges says:

    When I had first read about Hula, I was left with a bad taste in my mouth that I could not understand even has a I got excited about what I could do with it. Now I know. "Groupware", for me, has always been software that was beareaucrat-laden, bloated software that never did what I wanted it to do (at least not easily) and always tried to stick me into using its Magic Format^(tm). As I read more about Hula, the dark shadows of poor prior experiences flickered in the back of my head.

    As much as I hate to admit it, sometimes I have looked at software from the "will this get me (a 20 year old male college student)?" viewpoint. Hula stands as a piece of software that I would like to incorporate into a nice little server for my large, extended family to communicate and another little set up for close friends. Getting laid isn't high up on the list of possible "direct results" but, indirectly, I might just be getting lucky. *rimshot*

  16. So I said, narrow the focus. Your "use case" should be, there's a 22 year old college student living in the dorms. How will this software get him laid?

    This is the single best, or at the very least the most entertainingly succinct, summation I've seen recently of why so much software sucks so badly. I've always said that software should fundamentally improve a persons life in some measurable way (good calendars and good photo management sofware do this), but your way is much funnier.

    'Cause otherwise software just drives me to drink. Oh wait, being a lush improves my life. Never mind.

  17. Very good. My only quibble (as other's have alluded) is that real, honest to god, 22-year olds in cubilcs just want to book a fucking meeting room already and not have to get there and find someone else in it and dick around all afternoon trying to have the damn meeting when they could just be getting laid (or at least thinking about/planning it).

    That's the only _real_ killer feature in Outlook for most real people. But it's a doozy. Hopefully we don't need to bring in all the other crap to get it.

  18. roninspoon says:

    Here's the thing, you're so right. Those are all great ideas. As much as I hate to say it though, I'd love to able to use something better than Lotus Notes. Maybe I could use something that didn't eat a sack of dicks?

  19. tithonium says:

    Having my published .ics files parsed into an html calendar on my website hasn't gotten me laid (any more than I would have otherwise).

    Maybe I should pimp my evite clone more (needs a better name than 'Invitotron' tho).

  20. mattallen says:

    The funny part is that people that develop groupware read your blog as well... like me. Crap alpha groupware mind you.

    Now you've made me go back to scratch and rewrite the requirements, damn you.

  21. lifelike001 says:

    LMAO! id say the real challenge to the world is for someone to produce software that a solicitor can use. of the 30 lawyers i work with in a medium sized government division, theres probably half a dozen who can find the on/off switch. most of them use paper diaries and think groupware is a single unisex powersuit assigned to the division. please someone create a program that, when i send an urgent email to a lawyer, pops up on their screen in hot pink letters, doesnt let them do anything else until theyve read it, and possibly emits an electric shock through the keyboard if they dont!

  22. inoshiro says:

    "The first thing you want to do is make it trivially easy for someone to publish their calendar,.... Right now people can do that by publishing .ics files, ... If it's not HTML hanging off our friend's home page that can be viewed in any browser on a public terminal in a library, the bar to entry is too high and it's useless."

    I should point out, phpicalendar was a project which created PHP scripts which allowed you to plop little .ics files in a folder and magically have them appear as themed HTML anywhere on your personal site. There was a one-time setup you had to do, but it mainly involved little things like the URL, and if the week started on Monday or Sunday.

    For some reason, the actual phpicalendar site say it's "... willingly offline until further notice.," but you can still get copies of the source.

    I find it very handy to be able to sync my Palm to KDE, then export that to a website. I just wish there was an automation to keep note fields private (not exported), and have the calendar auto-exported on sync (rather than me generating an ics file every time). I suspect there is a Mac sync tool that does this, and that I could write a KDE sync plugin for it...

    • grafster says:

      You can set up iCal to publish your calendar to any WebDAV server, and phpicalendar provides that functionality in theory. Although in practice it doesn't seem to work all that consistently, unfortunately. I had it working once, and haven't got round to find out why it stopped working.

      • jwz says:

        "Any WebDAV server." As if that's something that actually exists in the wild.

        When I wanted to publish a calendar from Mozilla Sunbird, I saw that it wanted to use this "WebDAV" thing to do it, so I did some searching. What I found, mostly, was that the WebDAV people toot their own horn a lot without giving specifics; are very bad at describing what, exactly, WebDAV does; and do not give the impression that they have the first clue about security. (Maybe they do, but they don't give the impression that it has even been considered.)

        So I thought, "do I feel good about installing this crap in my web server?" and immediately answered myself "hell no", so instead I just push my .ics file out to the server with cron and rsync, and it works great.

        But -- my god -- it's not like this is rocket science.

        • jwz says:

          I do not mean to imply that mere mortals should be expected to deal with cron and rsync. Quite the opposite.

        • Exactly. Except I can't even do the poor man's work around... my ISP (Rogers Canada) turned over their e-mail and webspace management to Yahoo, and .ics files are not on their list of 'approved extensions' for uploading, FFS. I e-amiled them about it, and they pointed me at Yahoo Calendar, which of course only syncs with Outlook.

          I suppose it could be worse, Sympatico sold out to MSN/hotmail.

        • acroyear70 says:

          the other hassle with webdav is that its usually an add-on to the web server, not something that "any ole php script" can provide. thus, if you're using someone else's webhosting services, you likely can't get it without clearing it with them.

          the best standards, that allow the best interop, are those that *use* existing and supported standards, not those that *extend* them. consider the broad range of 3rd-party posting tools for livejournal. its possible because LJ doesn't do anything TO http; it uses it to POST content and puts all of the power into the content being POSTed. Much like how web services are so successful by just using basic POST and GET on XML instead of HTML; i don't have to add anything fancy to my server to get them (not to say web services don't have their own problems, of cource, but this is the IDEAL application for web service interfaces as opposed to the "stock ticker" and "metric to english" conversion scripts in all the stupid examples out there).

          WebDav, on the other hand, mucks about with all sorts of HTTP instructions and non-standard http headers that no "normal" tool has ever used or needed to. and does it at such a low level that its hard to get the tool in there and configured right, and then hard for the average person to control it once its working at all.

  23. moof says:

    To do this, they bought a company called Collabra who had tried (and, mostly, failed) to do something similar to what we had accomplished. They bought this company and spliced 4 layers of management in above us. Somehow, Collabra managed to completely take control of Netscape: it was like Netscape had gotten acquired instead of the other way around.

    For whatever reason, this seems to be a recurring theme: the bought company ends up borging and managing the buying one. It's happened at all too many places I've worked.

  24. geodog says:

    I hear your old Netscape story, feel your pain (I used to work with some of the Collabra people at GO), and think you can write a great story. But ...
    I've just spent the last month looking for a decent OS *lightweight* groupware product to replace MS Exchange + MS Project.

    There is a real demand for a decent product like that -- look at how BaseCamp is doing. Might not get you laid, but will get you paid + cred.

    • jwz says:

      Lots of things will get you paid.

      Most of those things won't make good open source projects, because almost nobody would willingly work on them unless they were getting paid to do so.

      • p3rlm0nk says:

        Not to dispute your point (which honestly I agree with) but what about I know I'd rather eat a sack of angry bees for lunch than worry about cornsilk blue linear regressions in spreadsheets, let alone coding cornsilk blue linear regressions in spreadsheets, but somehow people are working on that shit for free (and doing a rather good job of it judging from its appearance and functionality in recent builds). I guess it's a matter of Greater Suck: X might suck, but if X lets me flee Y which sucks more then maybe X doesn't suck so bad (coding office apps for free/ms windows. dialysis/being dead. etc.).

        • lherrera says:

          The driving force behind Open Office is a big company. They develop the less tasty pieces and oversee the whole thing. Geeks do just what they love to do, thus your "cornsilk blue linear regression in spreadsheets" is probably being developed by these guys.

          By the way, what the heck is a cornsilk blue linear regression in spreadsheets?

          • p3rlm0nk says:

            a sleep deprived fight club reference more than anything else. regressions are standard data analysis tools, and are frequently implemented in spreadsheets. probably the only thing more mind-numbing than re-re-re-re-re-re-reinventing the
            wheel is doing so for managers that care more about the widget's presence and color than what it actually does.

          • acroyear70 says:

            the other driving force behind OpenOffice is the fact that the hard parts were already done, and have been for years before StarOffice was bought and opened. the primary focus of the open-source edge to OO has been to increase its portability (major success), and increase its ability to process Microsoft files accurately (minor success, but ongoing).

            nobody is going to dig that deep into 5 year old spreadsheet code just to optimize another 2 seconds out of a large update or graphic display...but that's what PAYING customers will ask for.

      • geodog says:

        I don't think I missed your point, I just disagreed with it. I don't think that all groupware is about:

        "the chairman of the committee has emailed me this checklist, and I'm done with item 3, so I want to check off item 3, so this document must be sent back to my supervisor to approve the fact that item 3 is changing from `unchecked' to `checked', and once he does that, it can be directed back to committee for review."

        While some groupware is like that, some groupware is about getting things done as quickly with as few impediments as possible, and about communicating what you are doing and what you need. I totally agree with you about making them user focused first.

        Nobody cares about that shit. Nobody you'd want to talk to, anyway.

        Followed the "Getting Things Done" meme at all, and all the tools people are writing (without being paid) to implement it? Seen what people are doing with Ruby on Rails? Checked out BaseCamp, Tadalist or UseTasks? All attempts at lightweight groupware with elegant, minimalistic, user-friendly interfaces. Lots of geeks that I like to talk to (or read) are interested in working on this shit, but from the prospective of solving the user's problem, as you advocate.

        It seems like a least some people are willing to work on these willingly for cred or to scratch their own itch, as well as all the people hoping to make a business out of it.

        Now back to work, using MS-Project and Exchange (ugh)

    • zanfur says:

      I was hired at my current position explicitly to find and implement a replacement for exchange. An open-source drop-in replacement for exchange would be nice for that, but it's still a useless "groupware server". If I had to choose a single monolithic groupware server, I think has the right goals in mind (use what standards are available and don't bother re-inventing an MTA), but it's a pain in the ass to set up and synchronize with anything useful, like LDAP (or AD).

      As it is, I think the entire idea of a "groupware server" to be pretty counterproductive. Why bother with anything monolithic when there are so many "modular" (standards-compliant) ways of doing the same things? There are good calendaring solutions (anything html or iCal based comes to mind), there are good MTA's (postfix among others), there are good address book solutions available (vCard and ldap), and there are good shared folders solutions (well, if you consider imap "good" -- although my preferred method is through distribution lists). The only thing a groupware "server" does for you is give it all a single point of failure. The integration of email and meeting invites (i.e. calendar entries) is nice, but that's easily done client-side without any need for anything labeled "groupware".

      Shared folders are better implemented as distribution lists anyway (many will even archive it for you in some nicely html-exportable fashion) -- if you need data storage, use a frikkin' file server, that's what it's designed for. As for shared address books: LDAP. Done. Todo lists are personal anyway, and any sort of "shared" todo list is better stored in some sort of workflow or ticket tracking system ... which "groupware" just isn't. "Journals" and "Notes" ... yeah, personal. If you wanna share them, they belong in some sort of KnowledgeBase, Wiki, or Blog.

      Along the lines of <lj user="jwz">'s argument (I hope), what makes these things "good" is that they help yourself and anyone you need to work with. No one has to be "part of the system" to send or reply to meeting requests. No one has to be "part of the system" to get email from a distribution list, or view the archives of a distribution list. No one needs to be "part of the system" to look at an LDAP directory. No one needs to be "part of the system" to respond to email generated from a ticket tracking system. No one needs to be "part of the system" to look things up in a KnowledgeBase, Wiki, or Blog. One person using any of these is a boon to that person and those who work with him, and that fact becomes obvious, which causes new people to start using them. A groupware system only helps the people who are also part of the system, and even then only those that actively take advantage of it -- and it helps them less than a modular approach would, at that. A groupware system doesn't "slowly catch on"; it "slowly dries up" as it becomes less and less useful. Worst part of it is, you can't remove a useless aspect of one and replace it with a useful service.

      • zanfur says:

        Hula looks awesome, in concept anyway. I just wish it used straight smtp, instead of implementing its own MTA and message store ... or at the very least used something different than an mbox-type mail storage format. Incremental backups get killed by users with huge mbox's ... one piece of email and you back up the whole shmear.

        I don't think Outlook support will be much of a problem, come next release. I believe Redmond is integrating generic iCal mapi drivers into the next release of Outlook -- Outlook will be able to open any standards-compliant iCal file. At least, that's what the Outlook devs tell me.

        • I figured out the fix for incremental backups of large mbox or pst files. Install librsync, get the librsync perl module, hack it so it works with the new version of librsync, and make an app that transfers deltas for only the changed parts of the mbox or pst. You'll have to use cygwin to get it to work client-side on windows. (rsync does this exact same thing but it doesn't just transfer the deltas, it wants to keep a whole duplicate copy of the file on the remote side)

          MAPI drivers still require a working MAPI connector, of which there are no working open-source versions.

    • tangoglenn says:

      Thanks for the link to BaseCamp.

      I think there are developer reasons for wanting to have good groupware. But since the developers don't usually buy the tools, they don't get made from a developer's point of view.

      As a developer, I want the following so I can have more time to get laid:
      - I want to make sure that all of the things I need to do get completed for a release. I don't want to lose a weekend because something got lost on a sticky note that disappeared into my desk clutter.
      - I want the folks above me to know where I'm at, so they can leave me alone. Or, if they must speak with me, I want the questions to be along the lines of "what do you need?" not "where are you at?"
      - I don't want to waste time duplicating something someone else is working on.
      - If my plate's too full, I want to pitch things off it instead of digging myself into a hole because the work hero in me wants to prove how cool I am. I've never gotten laid for being a work hero.

      As a developer promoted to management because, well, working for someone else on the team would have sucked:
      - I want to have the "mechanics" of management take as little effort as possible. Spending a lot of time on GANTT charts so marketing can hype a feature that will actually make it into the product doesn't excite me. I'd rather spend my work time helping to figure out the development issues. Not quite as good as getting laid, but much closer on the fun scale.

      If BaseCamp doesn't solve it from the developer's point of view, I hope someone takes a swing at it.

      • geodog says:

        Kind of funny having a discussion about project management software on jwz's LiveJournal, but you did a very nice nice job articulating a set of requirements for one.

        Basecamp doesn't quite do it for me, feel free to send me email (geodog at cyberdude dot com) and I will be happy to send you a list of what I have found so far, or check out this guy's list

        There are so many attempts at lightweight open source project management apps that I think some of them must be people scratching their own itch, not just corporate hogs paying for it.

  25. edlang says:

    Wouldn't it have been equally as funny to see the Hula team jump through all the same hoops that you did?

    Funny in a thank goodness it isn't me, way.

  26. fastfwd says:

    God. This is so like my own experiences in corporate life. But the especially terrifying thing about it is that the business I was in is completely unrelated to software.

  27. thaths says:

    I thought the Collabra acquisition and the Groupware play was intended as a ploy to get attention away from the David v. Goliath Browser War stories. JimB, I think wanted people to perceive NSCP as a more grown ups company.

    I think we got the worst of all worlds in some ways - Marketoids from Apple, Middle Managers from Collabra and SGI,...

    • acroyear70 says:

      its, of course, far more complicated than that.

      unless you're microsoft with a 30% or greater return on investment (and not just a rapidly growing stock price, which is not the same thing by a long shot), you can't afford to just sit on your cash. microsoft can sit on their cash or spend it internally because nothing else out there can get the return.

      Netscape, on the other hand, got flooded with cash from the 1-2 punch of a successful product rollout and the IPO. with that, and with a (seeming) lack of ideas on how to keep the sales growing (the wall street curse: profitability sucks unless its eternally increasing profitability), acquisitions become the only way to increase sales and efficiency.

      of course, you can't make the case for an acquisition just on the relative profitability have to demonstrate the *synergy* (god, what an awful word) that will allow both companies to be more profitable by being combined (and not just because of the massive middle-layer layoffs that result, as today's mergers all demonstrate).

      very few acquisitions have any real synergy, any real meaning in the united company. most have some artificial pipe-dream of synergy invented by the execs to sell it to the stockholders. in the end, the lie behind it is often revealed (gee, it only took HP 28 months to figure *THAT* one out...).

  28. zhixel says:

    Great story, I have always wondered why Netscape 4.x was such a flaming piece of shit that was the total bane of my existance for a couple years there.

  29. remaker says:

    So when will you publish the book-length memoir of why Netscape failed? I'd buy it. You could even add a section on how to develop software that gets people laid. More people need to program with that ethic.

  30. crackmonkey says:

    After reading all that, I'm reminded of this :

    Still, I do like to hear your stories.

  31. jurph says:

    I work for a program office, and the first thing I did to improve productivity was to take myself out of all of the groupware spam lists. The next thing I did was start e-mailing and calling people for information instead of "logging in to the eRoom calendar roster scheduler" and using their "directed messenger" tool. Because webmail is only cool when it's actual webmail. When I have to log in and then navigate eight more clicks to get to an interface that's basically a broken webmail form... I have just wasted five minutes of your tax dollars, and also several processing cycles on a very expensive server that your tax dollars paid for.

    Groupware is right up there with Powerpoint for software that is killing our ability to do any real systems engineering.

  32. mark242 says:

    The day that Apache gets anyone laid, I will eat my hat.

  33. omnifarious says:

    This goes in my 'memories' for good and insightful ideas. :-)

  34. bitwise says:

    like I had just sprouted a third head

    Clearly you failed to inform us when you sprouted #2.

  35. moebius_rex says:

    Considering that I've worked in the "enterprise" software industry for, christ, about ten years now, this essay was pretty hilarious. Because, in a way, they're the folks responsible for the drive to make "groupware" out of just about everything.

    True enterprise software is about as unsexy as it gets, and will remain so forever, I think. This is because it's not made for people. It's made for hive minds. It represents the borg mentality jacked up to 11. And really, that's not a criticism of enterprise application design: enterprise software is basically designed to suit the world of suits that uses it, and in some cases it does this quite well. For that audience, it's all about the bottom line rather than the booty call.

    It's just too bad that the needs of the suits have to get mixed up with those of average joe noncorp individuals. But too many people design for them who hold the cash and are willing to spend it lavishly.

  36. divelog says:

    This is exactly what my company does. But it's all done the boring way with plain old html. Is that ok? I've also missed the jump from a calendering program to an evite program. Seems like messaging or filesharing is the next important step.

  37. So I was trying to do something with Lotus Notes. (Now you're probably saying "He's already fucked.") What I was trying to do was remove someone from my address book. You'd think that would be pretty simple, but I was having no luck. I spent at least a half hour going through Notes' chthonic "help" system. Complete waste of time.

    Then I used Google like I should of done in the first place and I came across the Lotus Notes entry in the User Interface Hall of Shame. It actually explains how to do it, albeit as an example of shitty user interface practice.

    Lotus should just replace the help system with a dialog box that says "Sorry. The Help system is as worthless as the rest of Notes. Please use Google."

  38. yesthattom says:

    It's an excellent essay.

    The one thing I would add to the calendar server is it's all about access rights. For a calendar to be really useful, I need to be able to give read-only access to certain people, read-write access to certain people (in fact, if I had a secretary, he/she should be given access rights to do all functions as me, without knowing my password). It's that kind of access that makes it really powerful.

    And get the users laid.

  39. go_team_ari says:

    I wasn't aware MA was a flyover state.

  40. thepef says:

    So, in the grand scheme of things, and considering I was a buyer of Netscape products at Mapquest back when. What happened with all the push visioning that was taking place at that time. By the way, I ended up working for Netscape, and then had the unfortunate happening of being sucked into AOL hell for a time being.

  41. collincollin says:

    I affirm that not a single curly brace escapes my fingers unaccompanied by a "Will this get my laid?". Oh and good work with this weblog, this was pretty well written.

  42. jwz says:

    One of the more entertaining referers I've spotted to this article: "Did he mention workflow?"

    Well damn. I get paid to do that kind of stuff. I bet that's what a lot of us get paid to do. And yeah, I hate to admit it, but it's fucking lifeless. It doesn't make anyone happy. Sure, commerce isn't about happiness (I guess), but it's not about life-sucking bureaucracy either. The word "enterprise" gives me a sinking feeling of sadness and grief.

  43. erg says:

    Of course it's funny, I don't know if their homepage reflected what you said, when you posted, but now? It's nearly word for word, what you suggested they focus on.

  44. spoonyfork says:

    For anyone who's interested, I wrote up some steps to getting a calendar published for free using Mozilla's Sunbird and