Why is Scheduling Still So Damn Hard?

Share via Twitter Share via Facebook Share via Linkedin Share via Reddit

Just talked about this with the folks from Zimbra (more on them in a later post), but the situation with scheduling is pretty much ridiculous. It’s 2005, and I can get satellite images of virtually any location in the world with a browser, but to schedule an appointment I have to exchange 3 to 5 emails, then manually create calendar entries. Nor can I get reminders of that appointment via the mobile phone infrastructure that’s been mainstream for what, a decade? This makes sense, how?

I was discussing this problem with Analisa Schelle from Ogilvy last week, while we were chatting on the business aspects of RedMonk, i.e. is it just us, or do we have help scheduling, etc? Unfortunately, given our (ok, my) fierce insistence on cost containment, it’s the former. Apart from some outsourced accounting, bookkeeping, editing, etc duties, scheduling falls pretty much on myself and my colleague. Normally this works adequately, in that we more or less manage to schedule and attend the meetings we need to, with a few dropped balls here and there.

But this process simply doesn’t scale. Think about how you schedule meetings with folks outside your own calendar system:

  • Step 1: Determine your own availability
  • Step 2: Communicate that availability to an external party; typically means cut and pasting or manually writing some openings into an email
  • Step 3: If you’re lucky, some of these work, and you receive a reply which requires you to create a new calendar entry
  • Step 4: If you weren’t lucky in Step 3, the available slots didn’t work, and the external party has proposed some alternatives so you’re back to Step 1. Rinse, lather, repeat.

I don’t know about you, but to me this process is crying out for innovation. Some of you are no doubt going to respond – what about Free/Busy? Outlook has it, Evolution has it, and it seems like it addresses at least a portion of the problem. And it does, sort of. The problem with Free/Busy, IMO, is that it’s distinctly non-intuitive within any of the major clients, hence no one uses it, hence it has little utility. Add in the fact that it’s read only, and I don’t think Free/Busy is the answer (is an average user likely to know what WebDAV is, much less how to configure their client for it?).

What I would love is to be able to offer – selectively or publicly – visibility into my calendar, with ACL based write access that’s easy enough for anyone to use. The sheer amount of time this would save come conference time, when 10’s or 100’s of people are asking for time with us, would be worth its weight in gold. It would allow us to push the burden of scheduling onto those who request our time. You want to meet with us? Great. Here’s the schedule, knock yourself out.

The current collaboration tools are so poor at this, however, that during OSCON I seriously intended to use our wiki as a calendar and scheduling tool. While that’s in part a reflection on how flexible and useful wikis are, it’s more an indictment of the state of collaboration today.

In many respects, scheduling is similar to the situation with web publishing prior to the introduction of blogging tools and wikis. At their core, blogging tools and wikis are nothing more than a dramatically simplified interface for posting content to a website and editing it. It’s not tremendously complicated, and not something that was terribly difficult before they existed – for non-technical users. Likewise, we have some primitive tools for sharing calendar information, but they’re not even close to being ready for Joe User – which limits their utility for everyone.

So to vendors or open source projects with interests in calendars, collaboration or scheduling, my recommendation is to solve for blank in this equation: blogging tools are to web content as ______ is to calendars.


  1. Why is time travel so damn hard? If only I could go back and buy Google stock at the IPO price I'd be rich today!

    But seriously, I think "blogging tools are to web content as ______ is to calendars" is not a realistic equation. Easy-to-update personal websites (blogs) are pretty easy given the existence of HTTP and HTML, and syndication (RSS etc.) was relatively easy because the basic semantics of a news feed were worked out by the wire services 70 years ago and could be easily mapped into XML. The fact that people fight over who REALLY was first to invent this stuff indicates than in a sense nobody invented it, or we all did maybe.

    But calendars are another kettle of fish … Not only are there no widely deployed formats and protocols that do 90% of what is needed and can be leveraged for this, the semantics are not well-defined (if I'm "busy" because of lunch plans with a friend does that mean I'm not "free" if the Boss has a flaming crisis? If I'm out of office traveling does that mean I'm not available for a teleconference via cellphone?) Likewise, blogs and feeds are public information; that's the whole point! But calendars are not … do people really want the world to possibly know about appointments with their psychiatrists, proctoclogists, etc.? Not that these are huge challenges to program around, but they DO require programming around and not everyone WILL get the privacy controls right the first time.

    This may be another thing I'm just cranky about because I (working at a group scheduling software company about 10-12 years ago) had thought that what we needed was a universal network, lightweight client-side scripting, and a data interchange format to make it easy to schedule doctors' appointments etc. We have the internet, Java/Javascript, and XML, and we don't seem a whole lot closer. Likewise this has been something was supposed to be a killer app for technologies such as the semantic web stuff. I'm not really saying that this would violate fundamental laws of physics such as time travel would, but I'm afraid that it is not among the low-hanging fruit. And I hope someone proves me wrong!

  2. c’mon, Mike – you expected something easy? 😉 i will certainly grant that in the above equation, web content is far simpler to manipulate than calendars, but seriously? you think it’s that impossible?

    maybe i’m asking for something undeliverable – your background in the under-the-hood mechanics far outstrips mine, but i can’t believe that at a minimum an improved and simplified free/busy (one usable by regular people) is technically impossible. that just doesn’t make sense. and at some level, aren’t we really talking about ACL’s for text files with defined and established structures? hell, just take iCal and make it easy to use, promulgate, etc.

    it’s obviously not easy – or we’d have it already – but it simply can’t be *that* hard.

  3. I said it’s *not* technically impossible, although the opening thing about time travel probably obscured that.

    I really would like to see something like what you asked for. I was mainly whining because I once thought about this for a living, had a similar vision, and thought the barriers were technological. The barriers are gone, and the vision doesn’t seem a whole lot closer.

    But my larger point is very much the same as my pushback on REST / AJAX / Web2.0 / etc. — Just because you can use the basic web technologies to pick the low hanging fruit doesn’t mean that that the higher-up fruit will come with just a little bit more effort. Once you have to wrestle with semantic mismatches, security and privacy issues, and integrating a lot of existing infrastructure that will not be thrown away to make it easier for me to schedule an appointment with the dentist or the mechanic, the problems are orders of magnitude harder. Not impossible by any means, and there are no alternatives I’m hawking that make the problem dramatically easier, but I’m not holding my breath. Again I really hope I’m wrong, and I have to admit that the scenario of my dentist or mechanic writing a blog would have sounded absurd 10-12 years ago too, so take this with mass quantities of salt.

  4. I totally agree this is the BLOG tool for the collaborative scheduling. What needs to be built is a standard way to export the calendar entries , I think iCal has a good shot at it and then some tools can be built to intermediate the negotiation. Another use case is where you don't really want to show all your calendar busy times but only show a view/subset of it. Anybody interested in collaborating on this one?

  5. Mike: well, at least we can agree that it's something that should happen 😉

    but in all seriousness, i think the problem may be framed incorrectly. what i'm advocating is a minimum-progress-to-declare-victory approach, not a once and future permanent solution. we don't necessarily need to have a seamless, beatiful solution tomorrow. i'd settle for usable free/busy as an interim step. then maybe limited editing of free busy, etc. then something more sophisticated. let's break the problem up into digestible pieces rather than try and solve everything all at once.

    Jobi: iCal would be my bet as well. i love the iCal facilities in Evolution; but making the process of exporting and sharing these is just too hard right now.

  6. No easy answers but thanks for elevating the questions!

  7. I think the biggest challenge is getting users (especially casual calendar users) into the habit of checking the schedules of those they wish to book meetings with.

    The desire to meet is usually initiated via an email that proposes the topic/purpose of the meeting and perhaps a suggested time. This triggers the back and forth of arriving at a time that works for everyone. Even assuming the technology is in place, to make the process efficient requires that user take the time to check the schedules of others… this is a change in human behaviour, which is always tougher to achieve than technology changes :-).

    [warning: blatent self promotion] I ran into the same issues that initiated this blog and setup a simple service for myself and my contacts to use. It hosts .ics and .vfb files and converts between the two formats, allowing Outlook and other client tools to at least have some interoperability. It will also convert .ics/.vfb into a visual representation of your schedule that you can send to others… i.e. “let’s meet to discuss this topic, you can see my available time at http://ifreebusy.com/neiljensen/freebusy” It’s not a true solution to the calendaring problem; but it is somewhat workable.

Leave a Reply

Your email address will not be published. Required fields are marked *