The 401st Blow :: Thoughts On Media

The iPhone Makes Apple’s Success With The iPod Look Anemic

Posted in Data Analysis, Technology by Noah Harlan on January 27, 2012

Apple announced its fiscal Q1 earnings a couple days ago and the numbers were staggering. There have been a lot of interesting stats people have noted. They had the second most successful quarter in all of business history – not Apple’s history – all business. Their profit – $13 billion – was more than Google’s revenue – $10.6 billion. Their revenue and profit were double Microsoft’s. And so on.

The driver of Apple’s comeback was the staggering success of the iPod. It was a device that changed how we listen to music, and it was (is?) ubiquitous. What’s amazing is that the iPhone has made that success look positively anemic. The iPod’s best year ever was 2008 with 54.8 million devices sold and its best quarter was Q1 2009 with 22.7 million devices sold.*

Q1 2012 saw Apple selling 37 million iPhones and 15.4 million iPads. Tim Cook, Apple’s CEO, said they sold 62 million iOS devices in the quarter. That means iOS is nearly 3 times more successful than the iPod at its peak. For the calendar year 2011, Apple sold 93.1 million iPhones – more than 2007, 2008, 2009, and 2010 combined – and nearly double the iPod’s most successful year.

*Apple’s Q1 is actually calendar Q4 (the holiday season). They start their fiscal year in October.

Tagged with: , ,

Spam vs. Hijacking

Posted in Data Analysis, Software, Technology by Noah Harlan on July 19, 2011

The most common and widespread form of spam is spam from an alien sender directly to me. Alien in the sense that they are unknown to me and directly to me in the sense that it lands directly in my inbox (or, more often, my spam filter). An email address like DrJosephAbudai@yahoo.cn and sent to dozens of variations of my email address. It’s carpet bombing and, like carpet bombing, is not very effective.

Increasingly the spam that does find its way into my inbox is of two related types. The first are hijacked direct sources: people I know who have had their accounts hacked and are unknowingly sending spam. And the second is indirect: social network accounts of people I know who have had their accounts hacked. Both types reveal a problem with traditional spam reporting. If I don’t know you and you send me spam then I am inclined to click the spam reporting link. If I do know you then I assume you’ve been hacked, but I don’t want to report you as a spammer because then you’ll potentially lose your account.

A solution would be a “report this as a hijacking” link adjacent to the “report this as spam” link on all messages. If the former is activated then the host would automatically change the passwords for the account, shut down outgoing mail, send an in-bound email to alert the user of the hijacking, and require re-authentication on the next login.

With this new feature I’d be inclined to report more of these incidents faster and networks & mail hosts could respond more quickly.

What Your Domain Says About You

Posted in Data Analysis, Technology by Noah Harlan on March 27, 2010

This is a very interesting data set from CreditKarma. They took a look at the email domain users had when they signed up for their service and crosshashed it with their financial information to come up with the follow chart of averages.

Via FlowingData

Brother Can You Spare A Dime

Posted in Data Analysis, Economy, Theory by Noah Harlan on March 25, 2010

I’ve noticed that ATM’s are increasing their prices again.

In the last few weeks I’ve encountered several ATMs that were charging $2.99 for a withdrawal. That’s an absurdly high fee. Or is it? If others think like me (a dubious proposition) it might be a loser for the banks. Here’s why.

When the fee at an ATM is $.99 I generally don’t think about the value of what I’m withdrawing. I’ll withdraw $20 or I’ll withdraw $100. I’m price insensitive at $.99 per transaction. I don’t think about the percentage of the withdrawal either and am happy to pay a 5% fee.

When the fee at an ATM is $1.99 I do think about how much I’ll take out. I generally will withdraw $100 or $120 if I have that in the bank. I will avoid withdrawing $20 as I feel like 10% is too high a fee. I’m price aware at $1.99 and prefer to pay a 2% fee.

When I see a fee of $2.99 I think a lot about what I’m doing. I wouldn’t take out $20 (for the same reason as if the fee were $1.99). I also wouldn’t take out $100 as I feel the fee is too high. In fact, I walked away from two of the three ATMs that asked for $2.99. This means I’m very price sensitive at $2.99 and unwilling to pay a 3% fee. But why did I use the 3rd ATM that charged $2.99? Because it let me take out $200 and I paid only 1.5%.

There’s one other thing to consider. It’s the time-value of money. Time-value is what money is worth to the person who has it right now. If I have $10 today to buy a house and invest, it is probably worth more than a contract that gives me $20 but pays it to me in 10 annual installments of $2. This is why you should always take the lump sum in a lottery payout if you win. The value of a significant sum now is greater than the value of a greater sum but doled out over a long period of time. Banks know this. They want to hang onto your money, even if only for a few hours, so they can earn interest and put that money to work. Earning interest on $1 for 24 hours doesn’t make much money, but earn interest on 20,000,000 customers’ $1 does.

Given this, perhaps banks would make more money by charging a smaller fee at the ATM.

1) They get the time value of keeping more since we withdraw less at each point.

2) They get a higher effective commission on the transaction since I’m willing to give up 3% when paying $.99 but only 1.5% when paying $2.99.

Or maybe I’m the only one who would think this hard about ATM fees.

Tagged with: , ,

Replying All

Posted in Data Analysis, Technology by Noah Harlan on February 10, 2010

When email first became prevalent in the mid-90’s one of the first phenomena to appear was the nightmare of accidental listserving. Everyone had that story of the girl or guy who was telling a friend about an embarrassing late-night hook-up but accidentally sent the email to their whole class’ listserver and lived in shame for the remainder of their time at that academic institution.  Well the problem persists, particularly in the form of “reply all”.

Today, on the wonderful blog Flowing Data (if you’re into data viz then it’s a must read), Nathan Yau does a great map of one of these conversations gone horribly awry.  What ensues if a Kafka-esque tale told in Reply All (and a sad look at the California DOT’s technical unsophistication).  Enjoy:

Cinematic Grade Inflation – Part II

Posted in Data Analysis, Theory by Noah Harlan on January 29, 2010

A brief follow-up on my previous post about Cinematic Grade Inflation. In that post, I noted that the average grade of the top 100 rated films (with at least 20 reviews) on Rotten Tomatoes had risen substantially from 2000 to 2009.

Following up on my post, Scott Macaulay over at the Filmmaker Magazine blog asked an interesting question:

I’d be curious to see the sample set of critics analyzed over the decade. I bet it’s a lot larger now, and I wonder if the new breed of critic is more disposed towards positive reviews than the critics we entered the decade with.

In an effort not to let that question be rhetorical (and out of personal curiosity) I decided to follow up and take a look.  What I found was not quite what I expected and, frankly, asked more questions than it answers:

So the average number of reviewers was higher in 2009 than in 2000, but if you look at 2000-2008 it actually appears to drop. Also, what was the reason for the extremely high average in 2004? And why so low in 2008?

Any ideas?

Data Underload

Posted in Content, Data Analysis by Noah Harlan on January 17, 2010

For Sunday morning – a nice, amusing little data viz via FlowingData:

Cinematic Grade Inflation

Posted in Data Analysis, Theory by Noah Harlan on January 4, 2010

Did films get better over the past decade or did critics just get easier?

Grade Inflation

When I was in college I was a computer science major. My school divided academics into three divisions: humanities, social sciences and science, and those of us in Division III all felt we had it the hardest when it came to grading. In science you are often graded objectively – things can be right and wrong – and when you’re wrong, your grade goes down. In social sciences and humanities the bulk of grading is subjective – the quality of your writing or your thoughts.  Our case wasn’t helped by one professor (clearly not a designer… jeeze Duane, that’s the same page you had when I graduated…) who felt that it was his personal mission to overthrow the grade inflation in academia. He graded on a bell curve and would post it on his door. A perfect semester saw him handing out as many A’s as D’s with the majority getting B’s & C’s.  It was frustrating, to say the least.  Now I work in media… go figure…

Rotten Tomatoes

Anne Thompson had a piece yesterday discussing the best reviewed films of the decade according to Metacritic and Rotten Tomatoes. The Rotten Tomatoes list got me thinking:

Rotten Tomatoes: Top 20 Best Reviewed Films of the Decade (minimum 100 reviews):

  1. 100% Man On Wire 141
  2. 98% Up 250
  3. 98% The Wrestler 205
  4. 98% Finding Nemo 191
  5. 98% The Hurt Locker 171
  6. 98% Let the Right One In 146
  7. 98% Spellbound 134
  8. 98% Chicken Run 134
  9. 98% Murderball 133
  10. 98% The Fog of War 133
  11. 98% Anvil! The Story of Anvil 122
  12. 98% The Band’s Visit 110
  13. 97% The Incredibles 224
  14. 97% Sideways 207
  15. 97% The Queen 177
  16. 97% Spirited Away 154
  17. 97% Once 148
  18. 97% Capturing the Friedmans 143
  19. 97% Maria Full of Grace 131
  20. 97% Winged Migration 125

In looking at the list, it felt like a lot of the titles were more recent films. Could it be that the list was unbalanced?  At first glance, no.  Ten of the films are from 2000-2004 and ten are from 2005-2009.

But when you look at it with two-year samples, instead of one-year samples, you see a slight preference for more recent films:

So, are films getting better, is there grade inflation, or is the sample set so small it’s neither.  I then took the top 100 rated films (with at least 20 ratings) for each of the last ten years and a pattern definitely emerges:

Something is up, and that something looks like the average grade.

The average rating for the top 100 films on Rotten Tomatoes has risen from 76% to 90% in the last decade.  That means ratings have risen 18%.  But let’s drop out 2000 as it may have been anomalously low. If we start in 2001 the ratings have risen from 85.7% to 90.3% or about a 5% rise.

So what’s the cause?  I see three possible solutions, enter your vote and let’s see what people think:

This Is The Right Time To Make Movies

Posted in Data Analysis, Distribution, Economy, Financing, Theory by Noah Harlan on May 31, 2009

This is a great time to make a movie.

There, I said it. I’ve been thinking it for a while but somehow hadn’t gotten around to putting it on paper (or in bits & bytes, as it were). Those of you who know me have heard my argument but I think it’s a good thing to say out loud, so say it with me:

This is a great time to make a movie!

I am not talking about creatively – though it is certainly a great moment creatively. Political and economic turmoil, changing technology, further democratization of the tools, declining costs of equipment (hello: RED, D90, GH1) all contribute to a fertile landscape and there are some tremendous advocates for the new models and methods of connecting to your audience. In particular I recommend following the smart work and writings of Ted Hope, Lance Weiler, Scott Kirsner and others.

But this piece is not a rah-rah, buck up and smile argument for creators (Ted’s been doing a great job of that). This is drawn out of conversations I’ve been having with people about why it is a great time to invest in media and content creation. While I believe this to be the case for all media and I am developing a number of cross-platform and transmedia projects, I am going to focus largely on film here. I believe the arguments I am laying out extend to other forms of content.

Let me also note that, though films can be made for virtually nothing, not all films should be made for virtually nothing. I love and admire amazing ultra-low-budget work like Shane Carruth’s PRIMER and Jake Mahaffy’s WELLNESS, however, like your choice of equipment & location, the method needs to match the intention (WELLNESS on IMAX is as absurd as BATMAN on DV).

Those films succeeded because they were coherently conceived to work at the budgets they were using and because of the intelligence, guile and creativity of their directors. That being said, sometimes you have to raise more money and when you do, you have to convince those you’re talking to why it’s a good investment. This piece is trying to address the question of why film is a better investment right now than it has been in some years.

It’s All About Timing

For the past twenty years we’ve seen an increasing number of films being produced and, more importantly, released. This has created an over-saturation of the marketplace that raises the hurdle for each individual film to succeed.

Most businesses have a saturation point for incoming capital. If you have a McDonald’s on three corners of an intersection then the law of supply & demand will likely not sustain all of those businesses. You have over-saturated the marketplace and a Darwinian contraction will take place. If you were to approach investors about opening a fourth McDonalds you would likely find it hard to explain the rationale for the investment. (Starbucks is a great example of oversaturation and contraction as elucidated by Lewis Black).

In a business plan for a traditional company you will have sections that deal with barriers to entry for your competitors, market demand and a void in the market that needs filling. Recently I was talking with an executive at a cosmetics company about a new line of products they were launching. She explained in remarkable detail each part of the market that was currently being served and then showed the specific niche that they were looking to step into. She laid out the rising demand in the segment, the other products that existed and what her company would bring that was different and compelling to the consumer. In essence, she explained how many people wanted her product and how FEW other people were providing it.

In film, our business plans have tended to put forward a nearly opposite argument. We have tried to show our investors how MANY films that are like ours exist and make money. We try to make our product seem as similar to other products in the market as possible (while maintaining that we’re unique enough to be marketed). If you’re making a small horror film you’re going to cite SAW and BLAIR WITCH and others to say to potential investors: “little films can make money so invest in a little film”.

This argument here can be defined as:

MARKET = DEMAND * PRODUCT

Where the product is your film, the demand is the audience’s prior desire for films similar to the product, and the market is the potential success of that film. We make the assumption that the product is singular and therefore its success is solely dependent on demand (ie: multiply demand by 1).

The problem here is that you have defined the market by removing the dimensions of time & quantity from the equation. If there were only a single film in the marketplace then this equation would work well, but what if there are many films? The demand should be split amongst them. In fact, the equation should be more like:

MARKET = DEMAND / PRODUCT

Where product is now ALL the films in the marketplace which meet the demand. We can agree that the distribution will not be even amongst all product but framing the equation like this can properly ratchet the risk.

But we still have a problem of time. If all films were released on a sequential timeline then this equation would work. This equation would describe a situation where we have 10 films, all horror, all budgeted under $1M, all released in one window (weekend). This is a saturation problem. We also need to account for the frequency of those releases since those 10 films spread out over a year might well work just fine. We focus so much on whether ANYONE will be interested in showing up that we don’t think about how many films that audience can show up for at once.

Our investment equation should really look like this:

MARKET = TIME * (DEMAND / PRODUCT)

Point 1: Film production does not respond to market saturation but film distribution does.

The Glut

I often describe the film industry as a thousand-watt bulb out on the porch on a warm summer night. Every moth for miles is attracted to it and they all will come crashing into it without much thought. The perception of glitz and glamour, the publicity, the myths, the artistic aspirations, the self-aggrandizing, the political posturing all contribute to that bulb and lure people from all corners into the industry. This means that when money is available, money will pour into the film industry.

Point 2: The production of film will expand to absorb all available capital.

The consequence of this is that there has been a glut of films produced and released over the last decade. In particular, while the number of studios films has declined slightly the number of independent films has grown substantially.

The number of MPAA & Indie releases per year, 1999-2008

The number of MPAA & Indie releases per year, 1999-2008

In 1999 there were 200 MPAA films and 256 independent films released theatrically (this includes foreign, docs, etc…). By 2007 those numbers had shifted to 188 MPAA films and 396 independents and in 2008 the MPAA released only 162 features while 444 independent films entered the theatrical marketplace. (note: the drop in MPAA features last year can partially be pinned to labor strife with the WGA and SAG). (source for all these stats is the MPAA)

This means we’ve seen an increase in the number of releases over all by 34% and independent releases of an astounding 73%. But when we factor in time we see how this becomes a tangible problem. The average number of independent films entering the marketplace per weekend has grown from 4.9 to 8.5. Factor in MPAA films and you have an average of nearly 12 films per week entering theaters.

Releases Per Weekend, 1999-2008

Releases Per Weekend, 1999-2008

The control point to pair that data with is the number of people going to the movies and that, unfortunately, has stayed nearly constant. In 1999 there were 1.44 billion admissions and in 2008 there were 1.36 billion, a decline of 5%. Though revenue has increased from $7.31B to $9.79B, that represents an increase of 33% which is roughly on par with the increase of total films (34%) but far behind the increase in independents (73%).

Point 3: We have been releasing too many films for the marketplace.

Putting aside the question of whether or not we can make 444 good films we have other problems. With so many films coming out each week it is increasingly difficult for each film to find its audience. Theater programmers know that there is a new crop of films coming down the pipe next week so if you don’t perform right away then you are pushed out by the wave of product behind you and there are more films competing for limited attention from reviewers, bloggers, and the general white noise of promotion.

Look at the case of the Israeli film BEAUFORT last year, which, the same week as receiving an Academy Award nomination, was pushed out of New York theaters to make way for new content. If a nomination can’t keep you an extra week then the marketplace is oversaturated.

You may now be saying: ‘but Noah, you’re only talking about theatrical and theatrical is dead.” You’re partially right. I am focusing on theatrical as those numbers are the easiest to find and break down but these issues extend to other mediums. We have to assume that each individual watcher is going to consume a finite and roughly constant number of films each year, regardless of platform. Perhaps they’ll watch a few more if they can get them easily at home, but that adds to the argument I’m making about the future. The place of consumption of content may evolve (TV, Cable, VOD, theaters, Internet, etc…), and the specific form of that content may evolve (featurettes, shorts, multi-part serials, ARG’s) but the consumption itself will still take place.

Furthermore (and I acknowledge this as an aside), I believe the theatrical market will continue to exist for a very long time and the reason has nothing to do with technology or cost. It’s about human nature. Teens want to get together, away from their parents but are too young for bars so they go to the movies. Parents want to give their kids to a sitter and go tune out for a few hours on a Friday night. People on dates don’t know what to say to each other and would rather sit in a movie and have something to discuss over dinner. The theaters are going to be just fine but we have to evolve our thinking about the economics of our content.

The Thinning Of The Herd

Since the number of films has been growing and we know that production expands to absorb all capital we can take a look at the production of films versus the stock market:

Releases of MPAA & Indie Films vs. the Dow Jones Industrials (DJIA)

Releases of MPAA & Indie Films vs. the Dow Jones Industrials (DJIA)

We can see that the expansion of independent films tracks pretty closely to the performance of the economy (as measured by the stock market). I will make the assertion that as new forms of media become monetizable that the cache of film will move across to the new media in a largely similar fashion.

What anyone who has tried to finance a film in the past nine months has encountered is the contraction of capital. The economic turmoil has contracted the amount of available capital and, following our axiom of available capital, that should lead to a significant contraction in the production of films. This will affect both private financiers (think: your uncle Morty the dentist) and independent financing companies who depend on wealthy private investors for their funds.

Point 4: Less free cash means fewer movies.

The Interregnum

Our industry managed to walk into two different buzzsaws at once. Before the financial crisis reached its climax in September we were already in the throws of our own economic turmoil. We were seeing the decline of one set of business models and the emergence of a new one, which we couldn’t model with any degree of accuracy.

As an industry we all know where we’re headed. Things are moving towards digital distribution. What we don’t know is what that model will be. You have ad-supported models (Hulu, Snag and YouTube), download to own (iTunes, CreateSpace/Amazon), online rentals (iTunes, Jaman), VOD (cable operators) and subscription services (Netflix) all competing to provide content. Ultimately we’ll see a combination of these services but the lack of consolidation and evolving pricing models make it hard to project the revenues for a given project. The only consistently emerging piece of data seems to be that the pricing model for a single DVD will not hold up in a digital distribution marketplace. Prices will be less and per-unit revenue will consequently be less.

But do not lose hope! We are also seeing the declining need for intermediaries for independent films entering the marketplace. Where once we needed sales agents to broker deals with distributors who would then broker deals with fulfillment companies who would broker deals with points of sale who would sell to the consumer, we now can sell directly and circumvent many of the middlemen. We can refocus our costs towards marketing and targeting our audience. The incremental cost of making your content available to a worldwide audience is crashing towards zero and this is a good thing.

Point 5: Digital distribution means a global audience for the same cost as a local audience.

Kevin Kelly’s idea of artists needing 1,000 true fans willing to spend $100 each is spot on but when you have a global audience you may be able to accomplish the same thing with 10,000 less-true fans willing to spend $10 or 100,000 casual consumers willing to spend $.99.

Where once the idea of reaching tens of thousands of consumers was an incredibly remote possibility for independent artists it now happens all the time, just look at YouTube. Jaime King got to 6,000,000 (by his estimate) viewers with STEAL THIS FILM and the incremental cost to him was virtually zero.

In the next 24 months the ability to monetize online revenue in a meaningful way will become a reality. I make this guess about timing by looking at the uptake of other related technologies in recent years. Consider that YouTube was only created in February 2005 and was ubiquitous by 2007. Netflix was created in 1997 and had shipped a billion DVDs by February 2007; they shipped their second billion in the next 26 months. Hulu went live barely one year ago and is now a top online video destination.

Whether it’s through iTunes, Hulu, Netflix or Vodo, it’s coming and the numbers will be substantial enough to at least make up for, and likely far exceed, the revenues from media distributed through traditional bricks & mortar channels with it’s production, packaging, shipping and storing requirements.

What we must do is to reorient our business plans to look at the remarkable moment we are in and how new content, professionally produced and financed, can have a more successful life than ever. A moment with this much possibility has not existed in the content world in a very long time.

The Key Points

So let’s summarize the key points at work here:

  1. Film is a risky investment. It always has been and it always will be. It is only a question of how to mitigate and evaluate that risk. (I say this because I believe if we are not up front with our investors we are bound to get into trouble down the line – also because some offerings require disclosure).
  2. The contraction of capital means there are going to be fewer films made right now. If I can make a film right now it will enter the marketplace with less competition than at any time in the last ten years.
  3. The cost of production is lower and incentives are better than they have been so your dollar will get you more than at any time in the recent past.
  4. The films being made right now are going to be entering the digital marketplace roughly in line with when we will expect the consolidation of that marketplace to take place. These films will ride the first wave of global digital distribution revenue. We have geometrically larger audiences with geometrically lower cost. The decline in per-viewer revenue is irrelevant.
  5. Whether you are making a traditional 90-minute feature or a ‘new media’ work, we are ALL in a new distribution model. As filmmakers we need to not cling to the arguments of past success but instead look at the future and show where our products can exist and thrive. (HT: Scott Macaulay for helping to clarify that point)

So get out there. This is a great time to make films.

Tagged with:

By The Numbers: The Rise And Fall Of Online Video And Independent Film

Posted in Data Analysis by Noah Harlan on February 15, 2009

**UPDATE** Some of the charts (the line charts, not the regional graphs) below were originally not stored locally (ie: I was referencing them on google) and, unfortunately, google’s charts seem to be evolving for no apparent reason. Thus, looking at these charts is slightly different than the charts used when originally composing this post. I’m going to replace the dynamic links with screenshots but my original data sets seem slightly different. I stand by everything below but it’s the charts are a little uglier.

There has been a lot of talk about the end of independent film. Conferences at Sundance and articles like this and this (often featuring Mr. Gill). But most are just postulating. Here are some interesting charts I put together with Google Trends just to see where things are headed.

**UPDATE** Scott Macaulay makes the very good point about units of comparison: “”Streaming video” and “online video” both refer to delivery mechanisms, whereas independent film is most probably being used to refer to a class of film. Certainly people searching for “online video” could be searching for ways to see independent film online. Or are you assuming that online video is referring exclusively to short form work?”

I recognize the problem Scott raises and that’s one of the reasons why I didn’t delve too deeply into analysis of the data. That being said, I am trying to track here how these ideas are related in popularity as memes and not as alternatives to each other. Scott is correct that two are delivery methods and one is a class of film but it’s not completely without merit to compare the interest in methods of delivery with interest in a class of film – and we see that as interest in certain methods of delivery have increased, interest in a certain class of film has decreased – that does not say that there is a causal relationship, only a loose correlation.

This first chart compares searches for the term “independent film” (in red) and “online film” (in blue). You can see the slow descent of independent and then, in 2007, the rise of online film.

"independent film" vs. "online film"

Independent Film vs. Online Film

Here is the same chart but this time adding in “streaming video” (in orange). You can see that in late 2008, for the first time, “online film” briefly overtook “streaming video”.

"Independent Film" vs. "Online Film" vs. "Streaming Video"

Independent Film vs. Online Film vs. Streaming Video

Here is a chart looking at search volume for “cinema” (orange), “film” (blue), and “video” (red). While I’m not surprised by the relative levels, and there is not much change in search volume revealing trends, what is fascinating are the spikes. It appears that ALL THREE values spike each year around December. I don’t have a reason why, other than that is when the better releases tend to be clustered and the year-end ‘best of’ lists, but it is interesting. Also, what was with the massive spike in late 2006 for “video”?

Cinema vs. Film vs. Streaming Video

Cinema vs. Film vs. Streaming Video

Now, if we dive into those numbers for the US we can see some interesting data. Below are US Maps, showing search volume, 2004-present, for the following terms:

“independent film”
"independent film"

“online film”
"online film"

“streaming video”
"streaming video"

Again, I’m not going to try to divine too much but the data speaks volumes by itself. The distribution of either interest or awareness (probably more the former but they are clearly related) is vastly more distributed for streaming video than for either online film or independent film. Now, to normalize that information for any difference between simply the words “film” and “video” here are those charts:

“film”
region_film

“video”
region_video

Interestingly enough, those terms seem to have relatively similar, and even, distribution (allowing that video is more heavily searched than film). So, there are some stats. I still think the feature film is not dead though.
**