Marco.org

Wrapping up the Democratic nomination

Wed, 31 Dec 1969 19:00:00 EST

It is, at last, finished. At least, most people understand that after a long, hard-fought race, Barack Obama has won the Democratic nomination. Most of the holdouts should come around when Hillary formally concedes on Saturday. Unfortunately, there are still a few rumors and loose ends that should be sorted out so nobody goes into the general election with the wrong impressions.

Toward the end of the race, the facts got a bit muddy as both campaigns tried to put a bright face on the situation. Unfortuantely, with all the competing messages, some of the facts got confused. In the end, people will decide on their own who to vote for in November—but I would like to make sure they have their information straight.

Did Barack Obama steal the election with superdelegates and insider support?

No. While the majority of the superdelegates ultimately backed Obama, Clinton led the superdelegate count until well into May. Obama didn’t catch up until after the Indiana and North Carolina primaries. The late deciders broke to Obama when the decided he was clearly going to be the winner.

If Michigan and Florida delegation had been seated with full voting rights as determined by their unsanctioned priamries, would Hillary have won?

No. If Florida’s delegation had a full vote, Clinton would gain an additional 19 votes. If Clinton had been awarded 55% of Michigan’s 128 delegates, and Obama had been awarded none (rather than the 29.5 the DNC gave him), Clinton would net about 100 delegates. (This scenerio would be insanely generous to Clinton because it assumes Obama got no support whatsoever in a state Jesse Jackson won in 1988). Because Obama had a lead of 127 pledged delegates, Clinton would still be behind in pledged delegates.

Did Hillary Clinton win the popular vote?

No. While the pledged delegate count is probably the best metric for popular support, a lot of people are interested in the nebulous and problematic “popular vote”. The biggest question on what to do with the popular vote is what to do with Michigan and what to do with states that did not report a popular total. There are, however, very good estimates of who voted how in those states that did not report a popular vote total—so the most inclusive estimate would use this. In Michigan, where late polls suggest Obama had as much or more support than Clinton, Obama was not on the ballot. Clinton won 55% percent of the vote. Nobody, or “Uncommitted,” won 40% of the vote. Write-in votes for Obama were thrown out. If we want to count Michigan, we should count the “uncommitted” votes for Obama. While a minority of the uncommitteds might have wanted to vote for Edwards, the 27,694 discarded write-in votes, the 20% of Clinton supporters who indicated in exit polls that they would prefer to vote for another candidate, and the disenfranchised Obama supporters who voted in the Republican primary or just stayed home should more than make up the difference. By counting as many of the votes in as many states, Obama narrowly but definitively wins the popular vote.

Was Barack Obama’s campaign financed by giant far-left celebrity donations?

No. It is true that Obama has broken fundraising records—but he follows campaign finance laws like everybody else. The guy who cut my hair told me he’d heard that Oprah donated $100 million to Obama’s campaign. Apparently he believed that Obama’s remarkable fundraising success came from a few big donors who somehow bought the primary process.

Under federal election law, the maximum donation to a political campaign is $2,300. Since the primary and general elections count separately, there is a maximum individual donation of $4,600. By law, any donation over $200 is public information, so you can check to see where all the donations came from. Oprah, for example, donated $2300 to Obama’s campaign. Obama raised over $264 million. Over half of that came from donations under $200. The donations came from over 1.5 million people. His fundrasing report was so massive that it broke Excel 2003.

Did the press and the pundits pressure Clinton to leave the race prematurely?

Not really. As Obama said, Clinton had a right to run as long as she wanted to. However, as early as mid-February, it became clear to anybody who did the math what was going to happen. The media may have sounded the death knell for her campaign before all the votes were cast, but if anything the extreme focus on states Hillary was likely to win helped the campaign continue long after victory became impossible. As primary contests go, the race was extremely close—but Obama still won comfortably.

Was Hillary Clinton more electable than Barack Obama? Will Obama lose to McCain?

Probably not. Polls suggested that both Clinton and Obama would beat McCain. Clinton does better in a handful of prominent swing states—but Obama does better in more swing states. Obama is also competitive in states like Virginia and much of the mountain west. November projections are premature at this point—but it looks like Obama is at least as well situated to win as Clinton was.

Did Hillary lose?

It’s better to say that Obama won. Hillary ran a very strong campaign. She was a very strong candidate. It was very close. A year and a half ago, I thought nobody would be able to beat her for the nomination. While there are always what-ifs, I think it is pretty clear that Clinton is the strongest non-Obama candidate the Democrats have put forth since… well, Clinton. None of the other contenders were even in the same league. Clinton had immense support and a massive number of votes. Obama just had more.

Old Media is choking on New Media

Wed, 31 Dec 1969 19:00:00 EST

I’ve been following politics pretty closely the past few months, and I’ve come to some conclusions about the media: there is not enough information on the Internet, blogs are too filtered, and instant communication hurts the distribution of information. This may surprise a few people.

My parents, knowing I’ve been something of a political junkie, frequently ask whether the lack of a connected television is difficult for me. What do I do during election night if I don’t have pundits in a box shouting at me? Of course, most of my information is coming from the Internet. Watching the pundit is a lot like reading a political website—except that you can’t skim the television. You can’t skip block quotes you’ve read elsewhere. You can’t cross-check anything that looks suspicious. If you want a specific bit of information, like what precincts in which county have reported, the Internet is a great place to look for it.

Unfortunately, the political reporters know this as well. While waiting for a campaign event, they are likely to be checking Blackberries to see the latest buzz on the gaffe of the day. The old media reporters are getting their cue from new media writers on what is worth pursuing. New media, in return, is getting most of its information from old media. Information (by necessity) removed from its context in one story gets further removed from context in another story. After reading political blogs, reporters ask candidates what is important to the arm-chair politicos rather than what seems relevant to the country or what nobody has asked before. Then, since communication is functionally instant, the loop repeats itself. Worse, since it’s easy to spend a large amount of time scanning many websites for little new information, there is not a lot of time left for actual reporting.

I like to imagine it wasn’t always like this. I like to imagine old newspapermen in their newspaperman hats would spend their downtime looking for a story by talking to people rather than browsing the Internet and trying to get in on a story that’s already been told. The Internet has great amounts of information—most of it repeated and respun multiple times. But it has nothing compared to the actual world. Rather than staring fixedly at a Blackberry, my mytho-newspapermen might have interacted with other people waiting for the same event and tried to get a sense of what they cared about and what was important to their community. If they wanted to see what other reporters were working on, they would have to wait for the next morning’s paper. They would be more influenced by their own judgment and hard work than the collective noise of their particular corner of the Internet.

In all probability, this rugged and uninfluenced reporter is a fiction. Worse, I probably got the image from a movie—so it’s not even my own fiction. And, in truth, the Internet is a great communications tool. But with all the glitz and appeal of new technology and new modes of transferring information, we can easily forget that good communication requires craft rather than volume or repetition. Good investigation requires time and subjectivity, both of which the Internet can steal.

I think this is part of the major harm of the Internet. It’s not web-addiction or cyber-bullying or MySpace stalkers or any of the other old, sensationalized problems-that-hurt-your-children translated into a new age. Rather, the harm is the difficulty we have sorting out what is productive and what is distraction. Whether it is a ball and a stick or a mouse and a keyboard, we’ve always had ways to waste our time. The Internet blurs the line between productive work and outright procrastination. This problem is solvable. Hopefully, in the next few years, we will sort out when it is time to disconnect from the web and experience the wide world.

Truth, rumors, and predictions for the next iPhone

Wed, 31 Dec 1969 19:00:00 EST

It’s very likely that an iPhone update is imminent. But while nearly everyone has agreed on that, nobody knows anything definite about the specifics: when will the update happen, and what will it change?

What we know

Apple’s WWDC conference begins on June 9. The new iPhone will most likely be unveiled that day during Steve Jobs’ keynote.
AT&T is prohibiting retail-store employees from taking vacations between June 15 and July 12, claiming that they’re expecting a “heavy selling period”. They implemented a similar policy last year for the iPhone launch.
AT&T executives have said numerous times that all of their smartphones will support 3G networking soon.
The Apple Stores have been out of stock of the iPhone, and the website has shown it as “Currently Unavailable”, for a few weeks.
The most recent developer-beta iPhone firmware contains a 3G toggle in Settings.
AT&T’s website briefly listed an “iPhone Black” as an option in a support form recently. They claimed that it was a mistake and a placeholder for testing, not a real product name.

And that’s about it. Everything else is speculation. Here’s mine:

Price and capacity

The current iPhone:

$399 for 8 GB
$499 for 16 GB

Apple usually doesn’t introduce new price points for the same products. Updates usually just replace the old products at the same price points. I expect to see capacity doubling:

$399 for 16 GB
$499 for 32 GB

It’s possible that they’ll also introduce a lower-priced model to increase sales volume. After all, Apple’s making approximately $20/month from AT&T per subscriber in the U.S. alone. They don’t need to make large profits on the hardware. If they chose to do this, it would most likely be:

$299 for 8 GB

I do not expect that the new iPhone models will differ in any other ways from each other — only capacity. I don’t think, for example, that they would only give a 3G radio to the largest-capacity model.

The new capacities and pricing would probably force a change in the iPod Touch lineup. I expect to see the same price points and doubled capacities in the iPod Touch lineup as well. (I don’t think we’d see a $199 8 GB model.)

Size

The iPhone is 45% thicker than the iPod Touch, but they share much of the same hardware. The additional thickness in the iPhone is probably to accommodate the cellular radio and a larger battery.

3G radios are larger and consume more battery power. Even if Apple finds a way to make the other components smaller, they’ll probably need even more room for the battery now — or they’ll have to disable 3G by default (likely). And they still need to find somewhere to put the 3G radio.

I don’t think Apple can make the iPhone significantly thinner yet, and they almost certainly won’t make it thicker, so I predict that it will stay the same size. (Depending on how much internal volume is consumed by the battery, the weight might change slightly, but we won’t notice.)

I also predict no changes in the physical buttons.

Hardware features

GPS: No. Not enough room, too much power consumption, not enough demand.
Better camera: No. It would benefit most from an autofocus lens, but tiny ones suitable for phones are still very rare and young. And they’re not tiny enough to fit in the iPhone yet.
Video camera: Yes. The current camera can almost capture video — it wouldn’t take much to add this capability.
Video chat: No. I don’t see them moving the camera to the front or adding a second one for pragmatic, size, and cost reasons, so this can’t be gracefully done.
Faster CPU and more RAM: Yes. Third-party applications and overall performance will greatly benefit from this. But these won’t be specifically mentioned, except maybe in blanket statements like “It’s faster.”
Battery life: Slight improvement, but nothing huge — less than a 100% improvement.

Software features

This will be the focus of the event. With the 2.0 software and the API release, iPhone users will be smothered in new software capabilities from Apple and third-party applications.

We will not see IM or MMS support from Apple.

Some of the enterprise-friendly features would make a lot more sense if .Mac got an update to support them: over-the-air calendar and contact sychronization, in particular. I expect to see this (most likely), or a partnership with Google or Yahoo to provide it (unlikely). I can’t see Apple only offering these with Microsoft Exchange support — why would they offer it to Microsoft server customers and not their own?

The IMAP spec contains rarely-implemented support for synchronized calendar events and contacts. Apple could use the .Mac IMAP servers to deliver calendar/contact sync by adding appropriate integration between Mail, iCal, and Address Book.

I predict that Apple will also add one more feature to .Mac that I haven’t yet thought of, and the combined new feature-set will be convincing enough that I’ll finally buy a .Mac subscription.

Disagree? Email me.

Comments

Wed, 31 Dec 1969 19:00:00 EST

I disabled comments on Marco.org tonight. It might be permanent — I haven’t decided yet.

Marco.org has had 263 articles published since 2003. Comments were enabled in March 2006, and we received 2,216 approved comments.

Most comments were idiotic. Many spewed ignorance, hate, sexism, and racism. Others were from confused people mistaking Marco.org for a Royal Dansk distributor, an LG Electronics support site, or a serious video-game news site. Nearly all of them were written poorly with endless language errors and bad formatting. A few were legitimate and well-written, but they were rare.

Almost none of the comments were worth the clutter on the page. And in a medium with infinite available page height, that’s saying a lot. (Resources might become scarce, too. I only have enough disk space for about 176,160,768 more comments.)

Doomed

Comments are almost always awful — they’re only tolerable when the audience size is tiny and you’re likely to know everyone commenting. But even Marco.org’s small audience is too big for that.

Comments are doomed to be this way by their design:

Audience: A commenter is guaranteed that anything they say will be read by a significant portion of the site’s audience.
Anonymity: There are no repercussions to saying anything stupid or negative because a commenter isn’t held personally accountable.
Invitation: Comments directly invite commenters to say whatever they want, usually with very little guidance and very few restrictions.

The filters that reduce people’s ignorance and aggression in the real world are absent from comments. There’s no reason to filter yourself.

“I want comments”

New blog authors often say they want comments, but what they really want is validation, feedback, and intelligent discussion. They assume that comments will provide this.

But they don’t.

Why reblogging works

Tumblr’s “reblog” feature is a way to satisfy much of the desire for comments. It allows other Tumblr members to quickly post a copy of someone else’s post on their own tumblelogs with an optional commentary or reaction.

This removes some the biggest problems of commenting:

Rebloggers are taking the discussion onto their sites, not mine. There’s less incentive to be ignorant or negative because they’d just annoy their own audiences, not mine.
Rebloggers are web publishers themselves, so they’re unlikely to be confused and mistake your site for Royal Dansk’s or LG’s. They’re also more likely to write well and understand what content is acceptable to publish.
The original author isn’t explicitly notified upon a reblog — the author must choose to look at reblogs. And it’s optional: authors can ignore reblogs and nothing happens.

It tends to generate exactly what people want: validation, feedback, and intelligent discussion. Clearly, then, these are possible to achieve online — just not with comments.

Quality and purpose

This is my site. It’s a direct reflection of me, and it represents a critical part of my online and professional identity.

I’m proud of the writing quality, and I’m always trying to improve it. Comments are a net loss. I don’t feel obligated to let strangers vandalize my site, in front of the audience that I’ve worked hard to build, for the possibility of occasional extended discussion or unintended humor.

I don’t write online for the miniscule ad money. I don’t care if my posts don’t appeal to Digg (in fact, I’d rather they didn’t). I don’t care if you think I’m an asshole, I’m an idiot, or I suck.

I write for myself. It’s fulfilling.

I don’t need anything from you to get value from my writing. My purpose is fulfilled even if nobody ever reads it. Do you have something nice to say, or would you like to intelligently refute my arguments? Thank you, that’s a nice bonus. Do you have something ignorant to say, or would you like to engage in an ad hominem attack? Don’t bother — it’s irrelevant.

There’s plenty of space on the internet for all of us to have our own places to publish our thoughts.

(You can support or intelligently refute this article over here if you’d like. It’s a discussion forum. Remember those? Your comments won’t appear on this page.)

Falling off the shoulders of giants

Wed, 31 Dec 1969 19:00:00 EST

Jonathan Tran wrote Stand on the Shoulders of Giants, an article arguing that software developers, and specifically web developers, should avoid replicating features available elsewhere. His argument is that we should build our applications on top of third-party services by the “giants” (e.g. Google, Yahoo) via their APIs instead of wasting time rewriting their functionality.

And he uses Instapaper, my read-stuff-later bookmarking tool, as an example of what not to do:

… the functionality of Instapaper is a proper subset of the functionality of del.icio.us. In other words, everything that Instapaper can do, del.icio.us can do and more.

Why then didn’t Marco build Instapaper over del.icio.us? It seems like a perfect match.

[…]

Instapaper’s value is solely in its amazingly simple interface. The reason I use it differently from del.icio.us is because its interface affords to different things. It makes different operations cheap, and that changes the way I think about those operations. But why doesn’t Instapaper integrate with my del.icio.us account, simply tagging things with a “read-later” tag? When it comes to bookmarking, del.icio.us is king; there’s no disputing that. Instapaper wouldn’t have less value if it were built on top of del.icio.us, it would actually have more.

For the sake of argument, I’ll assume that this is true, even though it isn’t, because Instapaper has features that Delicious doesn’t — and Jonathan seems to be severely underestimating the value of a good interface.

This isn’t a new argument — it’s the classic “build it vs. buy it” choice:

Reinvent everything yourself. This is usually chosen by those who don’t trust anyone else’s code, earning it the excellent nickname of “Not-Invented-Here Syndrome”.
- Advantages: You own everything, you know how everything works, and you can modify everything to fit your needs.
- Disadvantages: You must expend development time to build and maintain these components, and it’s easy to get distracted from your core competency with endless unnecessary projects because you won’t buy someone else’s bug tracker or invoice manager or shopping cart.
Use third-party software as often as possible.
- Advantages: Third-party software can be more specialized to a narrow task, and it might be better than what you could build in-house. If it breaks, it’s not your problem. New features may be added with updates at no additional cost. You can stop worrying about lower-level problems and jump right into the interesting parts.
- Disadvantages: Your ability to add or modify functionality is limited. If it breaks, you usually can’t fix it — you’re entirely dependent on the vendor. There’s no guarantee that the vendor will still exist to maintain the code in 6 months. And the vendor may decide to discontinue the product or change the pricing.

Like most religious programming arguments, there is no always-right answer: different situations justify different choices. But building Instapaper on Delicious would have been a terrible idea for many reasons.

Gains

Instapaper and Delicious do share some functionality: the storage and retrieval of a list of URLs. But that only took me a few hours to write. Nearly all other parts of the code (interface, user management, copywriting, etc.) took longer.

It would have taken me just as long to build an interface to the Delicious API than it took me to replicate all of the functionality I needed from it.

What would have been gained from that? Instead of maintaining the bookmark-storage code, I have to maintain Delicious API access. I’ve coded against a web-service API before, and good integration took a lot more work than I expected.

But there’s a significant cost to depending on someone else’s API.

Dependence

For one, not all of Instapaper’s users have Delicious accounts. For someone new to Instapaper, they would then need to confusingly sign up for two services. That’s enough of a barrier that most will say, “Never mind.”

And what about technical dependencies? If Delicious is slow or down, Instapaper would suffer or fail. Don’t assume that any service is invincible to downtime: a few months ago, Flickr was down for most of a weekend. And very few web-application APIs guarantee a level of service, so if I lost 8 hours of functionality and revenue from Instapaper because Deliicious was down, I’m out of luck.

There’s also no guarantee that an API will stay in its current state. What if the service discontinues the API because it’s too much of a maintenance or operational burden? Or what if they change the terms and it becomes unaffordable?

Competition

What’s the incentive for Delicious to even allow Instapaper to use it? While their functionality doesn’t entirely overlap, Instapaper is certainly a competitor. Why would Delicious willingly enable me to run a competing service on their infrastructure?

Innovation

Reinventing the wheel is often justifiable. If nobody had ever reinvented existing products, we wouldn’t have Google. Or the Playstation. Or nearly any Microsoft product.

Instapaper isn’t built on Delicious because it’s doing something different, despite sharing some underlying features. Sure, I could have built it on top of Delicious, the same way I could build a word processor with a set of macros for Microsoft Word, or Mozilla could have built Firefox as a GUI around the Internet Explorer rendering component, or Apple could have just abandoned Mac OS and made a pretty skin for Windows.

In each case, it just doesn’t make sense. The resulting products would suck and provide very little value.

If Instapaper were built on Delicious, it would be slow, complex, and confusing — it would lose all of the qualities that attract people to it. Instapaper has value not because of what it does, but how it does it.

Eight Belles and Thousands More

Wed, 31 Dec 1969 19:00:00 EST

After the filly Eight Belles suffered a fatal breakdown following her strong second-place finish in the Kentucky Derby, there are plenty of questions floating around. Since when do racehorses break down after the race? (Never before, in anyone’s memory, by the way.) Could something have prevented this? Some folks are just asking themselves why they watched the race, or why we race horses to begin with. Is it moral to ask them to try so hard that they kill themselves doing it?

Horses and riders risk their lives in every equestrian sport. If you’ve ever signed the waivers at a guest ranch or riding stable, you understand this. There are injuries, horrors and abuses in all sorts of horse industries. (Many of them blogged about most excellently at Fugly Horse of the Day.) That said, there are too many catastrophic, life-ending breakdowns in American Thoroughbred racing.

We race our horses to find out which horse is best (something many a backyard breeder never bothers to do), and in doing it, we have produced over the centuries a remarkably athletic and courageous animal. Thoroughbred racing is not going away, even if you stay home from the track and turn off the television. By being an educated and concerned fan, however, you can help make the sport safer for the horses and humans involved. And there’s plenty to be concerned about.

Our tracks are harder, our races are faster, and the horses we breed are more fragile than those in Europe. We’re working to fix some of that with the installation of Polytracks and other synthetic tracks around the country. Someday the industry will get its act together and end the use of steroids and other race-day drugs. I’m not saying Eight Belles was doped (though I’d love to know what was in her bloodstream, in the interests of full disclosure). Painkillers help get unsound horses to post, where they hurt themselves worse. And years of breeding horses who may have raced on steroids or painkillers means that breeders have not had available to them the most honest phenotypic expression of the genes that each stallion and mare has to offer. And speaking of breeding, I hope that after yesterday, despite the beautiful race that Eight Belles ran, breeders will think long and hard about sending their mares to Unbridled’s Song. Unbridled’s Song’s own foot problems helped keep him from winning the Derby in 1996. He likely inherited his speed and unsoundness from his Derby-winning sire Unbridled, and passed both along to Eight Belles.

Big Brown, the Derby winner, has been plagued for the duration of his short career by bad hooves, probably a legacy of his heavy line-breeding (that’s the polite word for inbreeding) to the Native Dancer sire line, the same one from which the Unbridleds come. A Wall Street Journal article from the day before the race points out that every horse in the starting gate had Native Dancer somewhere in its pedigree. All of that Native Dancer in Big Brown’s pedigree (most via his grandson Northern Dancer) may well give him the speed to be a superstar, as this pedigree analyst suggests. It could even make him a potent sire of speed. It’s also virtually a guarantee that he will pass along those bad feet and dancer-delicate legs.

The Thoroughbred industry has shifted in recent decades. Once upon a time, horses made money by pounding out many races over the course of long careers. Nowadays, breeders make money by selling young horses. Pinhookers make money by buying young horses at auction, getting them looking really good, and reselling them as two year-olds in training. Three year-old stars are ushered off to the breeding shed before, God forbid, something happen to them, because they are both too valuable and too delicate to keep racing. The emphasis on good looking two-year-olds and success in those early, shorter races and pre-sale breezes is why our industry is failing to produce sound horses, and why it’s failed for thirty years to produce a horse good enough to stand up to the rigors of the Triple Crown campaign and finish the job by winning at a mile and a half in the Belmont. The test is not too hard; we’ve just lost sight of how to prepare and breed the animals to pass it.

The darkest and most unconscionable thing about the Thoroughbred industry, however, is not the horrific on-camera breakdowns or the short-sighted breeding. The death Eight Belles died was swift and humane compared to what thousands of washed-up ex-racehorses suffer each year in slaughterhouses, often after traveling hundreds of miles, standing in cramped conditions on their injured legs. Or their perfectly sound, serviceable, but not-quite-fast-enough legs. It’s not only Thoroughbred racehorses, of course, but I’m sensitive to the argument that if you can’t rescue them all (save through legislative action), the Thoroughbreds who risk so much for their billion-dollar industry should damned well be immune from dying that way. Reading about slaughter may be the final straw to turn some of you off of Thoroughbred racing forever, after yesterday’s terrible loss, but for those of us who love the Thoroughbred and the sport, it’s a call to keep pushing for changes, and to support the organizations that help former race-horses realize their potential off the track. I think that some of the best advocacy for treating these horses humanely and giving them a second chance is done by everyday owners. As they ride off to second careers as pleasure and show mounts, they show the world that they all deserve that chance.

A few excellent organizations that help Thoroughbreds:

ReRun – A nonprofit that has adoptable ex-racehorses in New Jersey, New York and Kentucky.
CANTER – A nonprofit that helps find second careers for racehorses in the East & Midwest.
Old Friends – A nonprofit that retires notable or at-risk racehorses to a beautiful farm in Kentucky, where you can visit them.
TB Friends – A ‘mom and pop’ organization dedicated to rescuing and re-homing in California. Joe buys from killbuyer’s feedlots, from auctions, and takes in donated horses direct from the racetrack.

Unpaid internships

Wed, 31 Dec 1969 19:00:00 EST

These bother me.

The jobs that full-time students can get during the school year are usually low-paying, part-time retail or food-service positions. Many colleges are in small towns where nothing else is available even if students did have time to devote to them — but they don’t. And many students don’t work paying jobs at all during the school year so they can focus on their schoolwork.

Therefore, most students work as much as possible during the summer so they can accumulate enough money to last them through the school year.

When I was in college, I couldn’t afford not to get paid in the summer. And I was luckier than most — my mother graciously paid for my education. I still had some (relatively low) expenses, so I worked to pay them.

In New York (and probably other places), it’s normal and common for companies to offer unpaid internships. The interns are usually doing valuable, skilled, nontrivial work. Often, they’re picking up excess load from paid employees and doing the same tasks. They’re not just getting coffee and making copies… they’re doing real work. The same work that people get paid for.

But it costs money to be here, and this just helps contribute to the growing income and opportunity gap. The only people who can afford to take unpaid internships are the locals (so the student can stay at home) and the rich.

A middle-class student from Pennsylvania can’t afford to take an unpaid internship in New York: it would cost him thousands of dollars just to live here for the summer, and he’d lose the opportunity to make a salary at a paying job.

Why is it right to penalize those less wealthy in an educational setting? Why can employers bypass the minimum wage and get free labor by using a special word that often serves only to indicate the unpaid status?

Don’t say your company can’t afford to pay interns. We both know that’s not true. If their work has value, you can compensate them for it. If it doesn’t have value, why should you waste their time and yours? Why should anyone do it?

If interns are doing valuable work, pay them.

iPhone application pricing

Wed, 31 Dec 1969 19:00:00 EST

Third-party applications created with the iPhone SDK will be available for sale in June. Apple has created an incredible platform here — how much will these applications cost?

From a developer’s point of view, let’s see how much potential the market holds. This will determine what a developer needs to charge to make the effort worthwhile.

As of Q1 ‘08, Apple has sold 3.7 million iPhones. I can’t find numbers for the iPod Touch, but I’m sure the iPhone has outsold it many times over, so I’m not expecting it to contribute significantly to these figures yet. Let’s just assume that the Touch will negate the portion of those 3.7 million iPhones that aren’t being legitimately used — some have been lost or broken, some have been unlocked and won’t access the iTunes application store, and some will just never be updated with the 2.0 firmware required to run the applications.

Here comes the rampant speculation. If we assume:

1 in 1000 iPhone owners buy your application
You charge $10
You pay 28% income tax on the profit, and it’s all profit after Apple’s cut
Your support costs will be insignificant

…then you will get to keep $18,648 after tax.

Now, those are a lot of assumptions — especially the support cost. But the most important factor, and the biggest unknown, is the sales rate: will 1 in 1000 iPhone owners buy a decent application?

Move the decimal point in either direction, and it becomes very different: if you can convince 1 in 100, you get $186,480! But if you can only convince 1 in 10,000, you only get $1864… that’s not worth most people’s time to write.

And that’s an important factor: consider how long the application took to write. At 3 man-months, the 1-in-1000 figure above gives you $6,216 per man-month. Not bad.

But if you’re making a very complex application that takes two developers and six months to write, that’s only $1,554 per man-month — you’ll need to significantly raise the price or sales rate to make that worthwhile.

Now, over time, the number of iPhone owners is likely to increase significantly. Once Apple hits 10 million iPhone owners, the default scenario above earns $50,400.

The biggest and most important variable is clearly the sales rate. How many iPhone owners will use third-party applications at all? Among them, how many will ever pay money for one? Then, among those, how many will pay money for yours?

Considering this, I’m thinking that $10-20 is probably a fair price for most good, moderately complex, single-developer applications.

But it depends what everyone else does: if the majority of the third-party developers price their apps in the $5 range, and people refuse to pay for anything significantly more, many applications simply won’t be worth making. Conversely, if enough people are willing to spend $20-30 on good applications, they become much more worthwhile to write, and we’ll see more specialization and competition. The numbers could also be skewed if the iPod Touch becomes a major application-sales platform. (I don’t think it will.)

As a developer, I’m hoping people are willing to pay good prices. And I bet they will. I haven’t decided on my application’s price yet, but it will probably be in the $10-15 range. It definitely won’t be less than $10.

Don't Seat Michigan's Delegates

Wed, 31 Dec 1969 19:00:00 EST

Politicians can be myopic. Under current party rules, my state’s delegation will not be seated at the Democratic National Convention in Denver. All of Michigan is disenfranchised. Prominent Democrats are franticly searching for a way to seat these delegates in a way that doesn’t appear to compromise the ongoing primary process. They worry that not seating the Michigan delegation will leave Michigan voters bitter and hurt them in the November election.

They should give up. Michigan voters are already embittered—and fussing about how to seat the delegation without a meaningful election will only make things worse.

Here’s how we got into the mess we’re in. Strike that. Here’s how the Democrats got into the mess they’re in:

Iowa and New Hampshire get the first caucus and primary respectively. Why? A long time ago they decided that they get to go first—and nobody with political ambitions wants to anger them because, after all, they go first. This means that no Presidential hopeful wants say anything that angers either state—and questioning their primacy is an easy way to do that. Thus, the Democrats came up with a plan where Iowa, New Hampshire, Virginia, and this year, Nevada have contests before Super Tuesday and the rest of the states sort of wait their turns. Any state that didn’t play ball would would have all of its delegates stripped. Additionally, any candidates who wanted to be on the Iowa and New Hampshire ballots had to sign a pledge not to campaign or participate in any of these contests. All of the candidates did so.

Michigan thought this system was bad for Michigan and bad for America. (I tend to agree. And so did Florida, who did much the same thing and ended up in a similar situation.) They decided to call the Democratic National Convention’s bluff and move their primary to the very front. It was a bluff, right? The DNC could never be short-sighted enough to disenfranchise a critical swing state in order to uphold arbitrary and recently drafted party rules? Wrong. Never underestimate the DNC’s capacity for short-sightedness.

Michigan was stripped of its delegates. No candidates campaigned there. Most of the major candidates removed their names from the ballots. For some reason, Hillary Clinton did not, though she said (at the time) that the Michigan primary wouldn’t “count for anything.” Michigan’s move triggered automatic “we go first” rules in Iowa and New Hampshire, moving their dates up. Thus, despite Michigan being the third contest, it was mostly a beauty contest.

Voters could choose between Hillary Clinton, Dennis Kucinich, Chris Dodd, Mike Gravel, and, if they wanted to show up and vote for nobody, “Uncommitted.” By the time of the primary, Dodd had officially dropped out of the race. The results? Hillary Clinton won with an embarassingly narrow 55.3% of the vote. “Uncommitted” came in second with 40% of the vote. Kucinich got 3.7%. Dodd and Gravel split the remaining 1%. Write-in votes for either Obama or Edwards were discarded. Clinton supporters will point out that Obama and Edwards supporters pushed to persuade people to vote uncommitted. The effort was unorganized and unfunded. That 40% represents Obama or Edwards supporters who somehow learned that write-ins would not be counted and decided to show up to vote for nobody in an election they were promised would not count anyway.

As it turns out, the primary is (relatively) close. The exclusion of Michigan and Florida will give an air of illegitimacy to the whole process if it is close enough that the two states would make a difference. Of course, if party officials change the rules to allow Michigan and Florida to suddenly count, that will give an even larger air of illegitimacy, unless one candidate has a large enough lead that it wouldn’t make a difference. (Currently, Obama is leading by enough that he would still have a lead even if Hillary gets credit for “victories” in the two uncontested states, but it would cut significantly into his lead and leave him open to wrangling at a brokered convention.)

If Clinton counts her Michigan votes, the “popular vote” starts to look fairly close. Of course, this gives Clinton 55% of Michigan and Obama nothing, which is one of the reasons “popular vote” gets the mocking quotation marks.

By any account, it’s a mess. Leaving Michigan’s delegation out is a slight to Michigan—which the Democrats need to win in November.

Are there better options?

A second primary?

One option that got tossed around was the redo-primary option. Clinton’s campaign pushed this one. Unfortunately, a statewide primary costs millions of dollars. Michigan’s economy is doing terribly already and the state already paid for another. The state doesn’t want to gouge millions out of other programs to bail out the prodigal Democrats. Neither the state nor the national Democratic party wanted to offer the money. If they even had that kind of money, they needed to spend it getting other candidates elected. Since neither could agree who was at fault, both thought the other should pay, which meant neither would pay.

A mail-in election was briefly floated as a cheaper option. The campaigns would hire a private company to do the counting—which raised some theoretical concerns. But disenfranchising an entire state raises more theoretical concerns. The Obama campaign objected that the mail-in vote would not be “fraud-proof.” This was probably too mild. When people have to show up in person to vote, poll workers are there to spot somebody who is trying to vote multiple times. You might be able to vote in different precincts using different names or the names of the recently deceased—but overall, it would take an immense effort to garner (maybe) five fraudulent votes. With the mail-in primary, you could mail in as many ballots as you wanted using the names of as many people as you knew wouldn’t be voting and have a negligible chance of being caught. Alternatively, you could cast ballots under the name of somebody you knew would be voting against your candidate so that two or three ballots showed up in that person’s name and their vote would be tossed out. Then there are the more traditional forms of mail interception and tampering. One dishonest postal worker could wreak havoc on the entire system. So I guess you could call the plan “not fraud-proof.”

Perhaps a traditionally primary could happen if it were privately funded? Both campaigns have raised hundreds of millions of dollars. If they each ponied up $5 million, they could cover it and not miss the money. Alternatively, they could tap some of their billionaire buddies and privately fund the primary. A privately funded election should raise a few concerns about the integrity of the election, but whatever. The whole process is enough of a mess that we’re past worries about conceptual integrity. We just need something practically fair, reasonably secure, and representative.

Then who gets to vote? The January 15 election was an open primary. Anybody registered to vote could vote for either party. Some Democrats voted for Romney, hoping to give the Republicans the most odious candidate possible. Other Democrats voted for McCain, hoping to at least knock the most odious candidate out of the race. Can registered Democrats who voted for Republicans in January vote in the new primary? What about registered Republicans? The Clinton camp insisted on getting the list of who had voted for whom and excluding anybody who had voted for Republicans. Unsurprisingly, this would exclude quite a number of Obama’s supporters who either didn’t get the memo on voting uncommitted or didn’t trusted everybody who told them that the Democrats wouldn’t count Michigan and wanted to vote for the least odious candidate that would count their vote. One survey showed that 18% of Obama’s supporters would prefer Clinton to McCain. That’s a huge percentage of Obama’s supporters who could have taken a Republican ballot rather than vote for Clinton. Similarly, the on-the-fence independents (who traditionally have broken toward Obama) almost certainly also voted in the election that counted. Clinton’s supporters insisted that those who had voted for Republicans couldn’t vote again. Obama’s supporters would not back the plan. In fairness to both sides, it is unfair if some people get their vote counted twice, but it is also unfair if the rules change mid-game and those same people are suddenly not permitted to vote for their candidate of choice. In the end, the question was moot because the party could not get the list of who had voted in the Republican primary. (There is a history of litigation over these lists and whether the first election was even constitutional—but that’s another story.)

Maybe a caucus?

For a while, some Obama supporters were hoping there might be a Michigan caucus to award delegates. Washington has both a caucus and a primary—but the caucus is the one that awards delegates. Since a caucus is a different form than the primary it could partially avoid the double-voting problem. Best of all, a caucus is significantly cheaper than a primary. The turnout is lower. There is no secret ballot. It only lasts a few hours at a specific time of day. All these things save money. It also makes it a significantly less democratic system than the primary. Many state parties like caucuses because it’s a way to get a bunch of people out to a party meeting and get more people involved. It caters to the most enthusiastic parts of the party who are willing to give up a few hours of their day to support a candidate.

Clinton’s campaign nipped this idea in the bud. They said, “You can’t replace a primary with a caucus.” Technically, of course, you can. It would be a step backwards to replace a proper primary with a caucus—but the primary was far more flawed than even the most mangled caucus could be. (I’m thinking of you, Nevada and Texas!) What the campaign meant was that Clinton would lose a caucus like she has lost almost every caucus so far.

The caucus caters to the most enthusiastic—and Clinton suffers from an enthusiasm gap. She has frequently picked up the “late deciders” in elections.

Who are the late deciders? I’m a “late decider” every year in the Super Bowl. The Super Bowl’s this Sunday? Huh. Who’s playing? Ah. The Patriots. I’ve heard of them. Go Patriots… I guess. We “late deciders” watch the game if our friends invite us to a party. We eat the snacks. We cheer when it would be awkward not to cheer.

When it comes to an election, the late deciders go to the polls after they see a certain number of “I Voted” stickers and feel awkward without one.

Clinton’s supporters have as much a right to vote as anybody else, but Obama’s margin of victory tends to increase when the lukewarm supporters decide to stay home. So the caucus idea failed.

Can we just seat the delegates as assigned?

Not really. The election had a few problems. First, one candidate’s name was on the ballot and the other was not. Incomplete ballots might work in China or Iran—but not here. Secondly, neither candidate campaigned. There were no debates. There were no television ads. There were no rallies or chances to talk to the candidates. The only people with yard signs are those who imported them from other states or made them themselves. Michigan voters did not get a chance to evaluate their candidates. Thirdly, Michigan voters were told their votes would not matter. Many didn’t bother showing up.

Can we do something weird to get the delegates seated?

Recently we’ve seen a few novel proposals for how Michigan should assign its delegates. One Michigan congressman recently floated a proposal that would award some of the delegates based off the “primary” results and some based off the national popular vote. This would allow the party activists a chance to vacation in Denver and sit in on the convention. And technically it would seat the Michigan delegation. Or it would seat a Michigan delegation. But not anything that reflects the actual will of Michigan.

One of the Obama people has proposed splitting the state evenly. That would also seat a delegation. It would be fair to the two candidates. But it wouldn’t give Michigan a chance to vote.

These compromise solutions that try to seat representatives of Michigan without regard to Michiganders having a meaningful chance to vote for those representatives still disenfranchises Michigan’s voters. We used to have this sort of virtual representation. Then we had the Revolutionary War.

So what should happen?

Let’s be clear. We got into this mess because the Michigan Democrats broke the stupid rules and the Democratic National Committee decided to enforce the stupid rules. None of these proposals will be fair to both campaigns and the people of Michigan.

If we’re going to salvage anything from this mess, it should be a lesson: don’t disenfranchise an entire state.

Michigan’s spot at the DNC should be roped off and left empty. The state’s name should be read in the roll call and greeted with… silence. Where is Michigan? Oh. That’s right. We excluded them. Kind of awkward, isn’t it?

At this point, that awkwardness is the only representation I expect. The Democrats put an internal party struggle before the voting rights of the people. Admitting they screwed up is the first step toward reconciliation.

Leave the Electoral College Out of the Primaries

Wed, 31 Dec 1969 19:00:00 EST

I just read that Hillary Clinton might be winning in “the only count that matters.” Really? And what innovative accounting method is this? Apparently, Hillary would be ahead if we were counting the electoral votes in states she has won. This winner-take-all system is now the system that “makes sense” and Hillary should be winning. Let’s do ourselves a favor and kill this argument before it wastes anybody else’s time. We’ve seen a number of innovative counts to explain why Hillary’s campaign isn’t dead in the water—and this is the dumbest yet.

We keep seeing new electoral accounting methods because the Democrats have these shadowy figures called stupid-delegates (or something like that) who can tip the election one way or another. Many of them may be myopic enough to actually try to exercise this power—but enough of them are sentient enough to realize that party insiders thwarting the will of the people will look particularly undemocratic and would hurt the winner in November. This group will vote for the winner of the actual contests.

Counting

That’s where the funny counting comes in. Obama has a commanding lead in pledged delegates. Obama has won more states. Obama even leads in the poorly counted popular vote.

This is not the first novel counting method we’ve seen. For a while the Clinton campaign talked about how she was winning “the big states.” It was a shaky story because it counted uncontested Michigan, uncampaigned Florida, and home-state New York while discounting Illinois. Then it arbitrarily drew the line between Massachussetts and Georgia. And it counted the Texas primary but not the Texas caucus. Even more perplexingly, it did not explain why one “big” state should count more than a couple medium-large states.

Then we heard about the “states that matter.” Apparently states that routinely voted Democratic mattered—but Republican states did not. Perhaps the implication is that the Democrats should only care about the states that vote for them? Or perhaps this one wasn’t very well thought out. Sure, Texas is likely to vote Republican. But does anybody seriously think Obama can’t beat McCain in Massachusetts? This metric was quickly retired in favor of something less offensive to half of the country.

We heard about how Hillary is polling higher in the swing states like Ohio, Pennsylvania, and probably Florida. That sounded pretty credible until somebody counted other swing states. Oregon, Minnesota, Iowa, Missouri, Virginia. Additionally, Obama won by such great margins in some of the red states that they too might be in play if he is the nominee. As a small example, Obama received over three times more votes than McCain in solidly-Republican Alaska. It could happen.

The latest count is who would be winning in the electoral college. For the sake of the national dialogue, this argument should die an early death.

The electoral college

Democrats haven’t forgiven the electoral college for Gore’s 2000 loss. It’s a hard sell to argue that the bizarre-yet-Constitutionally-enshrined system of the electoral college is a more democratic system than the relatively proportional system in place today.

Sean Wilentz writes, “Democrats in primary states choose their nominee on the basis of a convoluted system of proportional distribution….” I’ll grant that the system is convoluted with open and closed contests, primaries and caucuses, and the mysterious superdelegates with the voting power of 15,000 mere mortals—but what is so confusing about proportional representation? That’s a bedrock principle of democracy, not, as Wilentz writes, an “eccentricity in the … Democratic nominating system.”

The current system started with the one-Democrat-one-vote ideal. It’s far from perfect—but it’s much more progressive than the winner-take-all-regardless-of-margin system that the electoral college offers. Superdelegates might accept a gigantic step backwards on the grounds of practicality—but it’s certainly not any sort of moral high ground.

A bare win in a state Democratic primary has little to do how many electoral votes that candidate is likely to win in November. The relevant question is not who won California on February 5, but who is most likely to win California in November. Clinton won California by a fairly narrow margin in February. But California will almost certainly vote for either Clinton or Obama over McCain in November. So will New York. Similarly, neither Clinton nor Obama is likely to win in Texas. Thus, for electability purposes, New York, Texas, and California are not the three most significant states. They are functionally irrelevant. In contrast, the mountain West in which Obama has won by gigantic margins has recently elected a spat of Democratic governors. Perhaps this block of states would be in play.

We can understand the electoral college as a compromise system. Each of the states decides how to award its delegates. To get maximum impact, all the states except new Hampshire chose a winner-take-all system. In the general election, the unit apportioning the delegates is the state. In the primary, it is the national party. The Democrats chose the system best for the national Democratic party, not the system that maximizes the influence of individual state parties. They are selecting the candidate with the most support across the country. To do anything else would hurt them in the general election.

Yet the biggest reason that all of these creative methods of determining the leader are counterproductive is that both campaigns were given the same set of rules in the beginning. They proportioned their resources according to that system. Were the rules different, the campaigns would have campaigned differently. Obama might have skipped New York entirely and focused on California. Obama’s campaign might have written Ohio off entirely but won Texas. We don’t know how things would have gone in an alternate reality where the election happened differently. We do know who is doing the best with the rules as they are established. Perhaps we can extrapolate from that. In the meantime, let’s leave the talk of the electoral college out until sometime when it’s relevant.