How baseball’s tech team built HBO Now | Hacker News Reader

How baseball’s tech team built HBO Now

theverge.com

103 pointscoloneltcb47 comments

The funny part to me about this story is the fact that HBO had to do a 3.5 month sprint to get this out on time. Seems like incredible incompetence that a company of their size could be this unprepared for such a major and obvious business move.

The even funnier part is that HBO spent years and millions on their inhouse .NET solution that was headed by a former MS exec who brought in a bunch of his MS buddies.

http://arstechnica.com/business/2014/12/hbo-cto-resigns-as-p...

>Anonymous sources speaking to Fortune say that the scrapping of project Maui was due to a growing distrust of Berkes. Berkes joined HBO in 2012 from Microsoft and built a Seattle office with 55 engineers, many of whom he worked with at Microsoft. The office’s separation from New York-based employees and its cost to the company apparently created divisions within HBO. Fortune wrote:

"According to sources, Berkes had known about a “memory leak” [on HBO Go] for nine months but decided it was a “non-issue.” That leak eventually led to the HBO Go outages. Internally, some accused Berkes of using the outages as a way to ask for more money to invest in his Seattle engineering team. He got the investment, but HBO executives have not been pleased with what he’s delivered. Berkes delayed product launches and was unable to deliver on upgrades. “If you look at what [HBO Go] is today versus two years ago, he hasn’t really done anything,” one source said."

> inhouse .NET solution that

This is not true.

Which part? That it was .NET?

The real question is why both HBO Now and HBO Go exist. They're almost the exact same freaking service. I honestly don't know how they're different, except one comes free with cable and the other only works with Apple devixes.

One quick point is that now HBO Now is available on non-Apple devices.

https://order.hbonow.com/#devices

Uhh that's already well understood. HBO Go was internally written and not very good. They contracted MLB to write a replacement. HBO Go is on a ton of platforms. There wasn't enough time to writ HBO Now for all of them in time for Game of Thrones. So they started with Apple. It's already on several non-Apple devices with more coming. They will keep adding platforms and eventually phase HBO Go out. It's pretty straight forward.

I'd say it's understood from a technical perspective, but it makes no sense from a customer perspective. Why have two apps that do nearly the same thing, and are named nearly identically?

HBO Now could have been another log-in provider for HBO Go. Alternatively, why can't I log into HBO Now with my HBO Go creds?

(Edit: This, to me, is a classic example of shipping the org chart.)

I agree, as a consumer the whole two HBO app thing is super confusing. This article is the first explanation I've seen and now it makes total sense; yes, they clearly shipped the org chart. (Another great example, on iOS, is that there are two Merrill Lynch apps with basically the same exact features and UIs).

Since a user typically only has to log in once in a while, it looks like HBO's gambit of adding lots of friction to the sign-up process vs getting video shipped really paid off.

Different products. Different billing. Different back ends. Hooking HBO Go accounts into HBO Now would require even more work. Which given there was a crazy rush to release in time for GoT doesn't make a lot of sense. And HBO Now only launched in North America while HBO Go is worldwide.

You aren't wrong that it isn't confusing for users. It won't matter in 2 years when HBO Go is dead. Everything is focused on getting to that point sooner rather than later.

You can't log into HBO now with you HBO Go credentials because they are two totally separate business models. You need a pay-tv subscription to authenticate HBO GO, while you pay a la carte for HBO Now and don't need cable. Cord cutter versus traditional.

Consumers don't care.

I agree that they are two different business models, but the content on both of them are the same, right?

Wouldn't it be easier to have two separate login pages on the same App? (One authenticates with HBO itself & the other with Comcast/Time Warner et al.) Once authentication is done, you get to see the same content.

They could have removed HBO Go on Apple's devices when Now launched & then gradually remove Go, as Now spread to remaining platforms.

> They could have removed HBO Go on Apple's devices when Now launched & then gradually remove Go, as Now spread to remaining platforms.

This sounds like what they should have done. Replace the app with "HBO Now", but keep the HBO Go branding, app store presence and icon. Make it a major version update.

You'd be surprised how deep into the stack login goes on these content-controlled A/V apps. I can easily see that a distinct login/billing requirement would lead pretty quickly to totally distinct apps.

On Android, HBO Go is way better than Now. HBO Now Android version is horrible, with terrible closed caption font, forced landscape on Nexus 7, no Chromecast support, just to name a few.

Any sources on this?

The NHL partnership is definitely going to be good for the game. Finally going to introduce some modern, quality, team websites in addition to potentially couple second delay livestreams instead of the current 90+ second GameCenter Live offering.

NHL is also toying with more tracking technologies [1] that could make for a pretty cool experience while watching the game live (The cost of pucks with tracking chips were ~$200 each, which is obviously too expensive for how pucks are used currently.)

[1]: http://www.nhl.com/ice/news.htm?id=750201

Now just give me NHL streaming on a per game pricing structure where I can pay to watch on demand and not be saddled with a $160 bill when my desire is to only see a handful of games over the whole season and we'll have a solid winner here.

http://www.broadstreethockey.com/2015/6/15/8783249/nhl-tv-se...

Its not a perfect system, but some per-team pricing is nice.

I've been an annual subscriber to the NHL and MLB services for 4 years. The NHL service has been awesome, the MLB service has quite a few problems and I find myself constantly wishing the MLB people would just do things the way the NHL does.

The NHL removes commercial breaks, MLB leaves them in. The NHL player has a way to skip 10 seconds forward or back - MLB player only does back. But the worst, the MLB streams constantly get stuck and then the stream jumps back to the beginning of the game. Then I get to start hunting for where I was.

Uhh, I guess Rogers GameCenter Live was a bit different. Rogers gave customers a free access for the whole season, which resulted in ads still being there.

I've actually had the same issues happen in GameCenter Live. Lots of glitches, laggy, and the quality was horrible, the actual game was much smaller than the full screen.

One annoying issue was when I wanted to watch the game well after it had started, or even after it ended but didn't want to know the score. I always had to put my hand over sections that showed the score until I pressed the "Hide score" button and then always quickly forcing it to start from the beginning instead of mid way through so I wouldn't see the score.

The league minimum salary (per player) translates into $6,400 per regular season game, I guess spending $10,000 or $20,000 a night on pucks would merely be irritating.

(teh Googles suggests they actually use less than 50 pucks, so $10,000 would be highish)

During the game, it definitely wouldn't be an issue. But many players like to give away pucks to fans, keep them after scoring important goals, etc... The NHL doesn't want to destroy the ability to hand out pucks.

Awesome, now I understand why there are two different HBO apps and why HBO Now is a little clunky.

Little confused about how the article is spinning BAM as a plucky underdog. They got $77 million dollars from Major League Owners, demanded another $10 million from TicketMaster for selling baseball tickets and have a monopoly on some of the most popular content on the planet. I'm glad they've solved some hard technical problems and particularly glad that I can watch Game of Thrones without a cable contract but not sure that they are a shining example of plucky little startup succeeding despite the odds...

Where in the article did it imply they were an underdog? They clearly spelled out the numerous advantages BAM has had, including the capital infusions, which implies the exact opposite of them being an underdog.

They're blatantly proclaiming BAM as being a rising juggernaut in the field of streaming, rather than an underdog. There is frequent promotion of their scale and accomplishments, including their revenue and how they're powering other large businesses.

Perhaps I'm mistaken in my reading of the article but the quotes below set a tone for me of an organization that was succeeding where it shouldn't have instead of an organization with huge financial backing delivering products that it was supposed to.

The group they assembled wasn’t a Silicon Valley dream team, more like tech’s version of Bad News Bears: a minor league beat writer, a former state treasurer, and assorted AV nerds from around the league.

BAM began as the in-house IT department for the league’s 30 teams, a small handful of employees originally tasked with building websites for teams and clubs. But over the last 15 years, BAM has emerged as the most talented and reliable name in streaming video, a skill set suddenly in very high demand.

That's talking about how they began, 15 years ago. I think the author was trying to get across that in the beginning, BAM was not a technical leader in their field. But as they faced large technical and infrastructure problems before other groups did, and they decided early on to keep development internal, they became a leader.

One thing the HN audience might find particularly interesting. BAM is owned by the league, and thus the teams and their myriad web of small owners, some of whom see their ownership of BAM as more valuable than their team.

I like HBO Now, but the video quality (over fiber, no less) is absolutely terrible. It's easily the flakiest thing on my apple tvs, as well.

You'd think playing video on a desktop computer is a solved problem, but HBO Now manages to make it stutter like it's the '90s.

Their code is doing something ridiculously inefficient with the CPU.

I've noticed numerous technical issues with HBO Now while watching on Apple TV.

There's the occasional frame rate drops from poor encoding, the frame rate should be constant, usually around 24fps for film. The audio levels are encoded ridiculously low compared to other services like Netflix.

The CDN itself seems to be over-saturated most of the time resulting in drops in bit rate or occasional interruptions. This is especially true on Sunday nights and doesn't matter how fast your connection is.

The highest quality (when you can get it) is 1080p but it doesn't matter because the bit rate must be no more than 5Mbps, maybe even 4.5Mbps.

There are numerous bugs in the UI, just the other day I selected the Series menu and it kept loading movies below it instead. The UX is horrible, it doesn't track what you've watched and there's no way to view recently watched. You can add an episode to your watch list but not a series so when you go back via the watch list you have to way of navigating to the rest of the episodes.

I also think the price is too high, $15 a month is just a little hard to justify based on the amount of content available. I see it more as a service I would just subscribe to a few months out of the year.

Considering the amount of devices you have to write streaming clients for these days (iOS, Android, Chromecast, Roku, PlayStation, Xbox, SmartTVs, etc, etc.), why not just partner with Netflix and offer it as a Netflix add-on using their device clients and CDN?

Realistically it's worth about $5-6 a month and if one could just add that to their Netflix bill easily then I'm sure they would get a ton of subscribers.

So, way back like few years ago, when Reed Hastings was asked who his competition is, he replied HBO. True to that Netflix got into the business of new shows. So the executives of HBO will never tag along with Netflix, that is why HBO shows are on Amazon Instant Video, not on Netflix. Even though for the end-user there is a level difference in service provided by HBO and Netflix, both these companies view each other as rivals.

I remember when Ted Sarandos (head of content acquisition at Netflix) said "The goal is to become HBO faster than HBO can become us."[1]

[1] http://www.gq.com/story/netflix-founder-reed-hastings-house-...

>Because if HBO Now had come out and face planted, there would be a lot of naysayers who turned around and said, ‘See, this is why the adults need to handle television.’"

I think "the adults" have proven they're completely out of touch with the average consumer and what they want. I think even with outages, people would've continued supporting and championing HBO's "cutting the cord" offering.

I'm rather surprised that there was no other live-streaming video team that could step in here. Can Netflix really not do live streams with only ~90s delay? Can't Hulu stream live? YouTube has had live streaming for a while, right? Clearly MLB (and other execs) want to own the Whole Product, but could they not have contracted this work rather than build BAM?

I appreciate that sporting events may have special technical needs (the World Cup was insanely popular), but I'm curious as to what special strategy BAM uses. Do they make special traffic deals for certain markets? Or perhaps are they really the first team to try to do very large-scale streaming with only a ~90sec delay?

I really wish Netflix would pursue live streaming sports, they certainly seem the most apt for the job.

A significant challenge is not necessarily live streaming the game, but the production. Most of the productions are ran by regional or national TV cable. They certainly don't want to encourage moving off of cable. Rogers (i.e., Sportsnet), who own exclusive NHL broadcasting rights in Canada (with exception to a few NHL clubs and some regional restrictions), are running NHL's Game Center Live in Canada. I don't know if the technology/provider of live stream is the same as the american version (non-Sportsnet) but because they own all national rights, there are no blackouts.

The U.S., version of GameCenter Live is pretty restrictive to geoblocking.

I don't think the roadblocks are in the licensing agreements that are preventing widespread live streaming.

http://www.broadstreethockey.com/2015/8/4/9094733/its-time-f...

According to the article they were doing live streaming in 2002, years before any of those.

Netlix - 2007

Hulu - 2008

YouTube - 2010

Hulu is US only, as far as I can tell, and Netflix still has spotty international coverage.

From CNet in October 2001:

"Next season, the professional baseball organization plans to begin streaming real-time video of games over the Internet, according to the head of its online division.

"The move, which will be targeted largely at employees with high-speed Net access at work, would represent the first time a major sports league has made live games available in their entirety online, according to an MLB representative."

http://archive.is/20120711061344/http://news.com.com/2100-10...

Why didn't they contract the BBC? iPlayer works insanely well for live content with live rewind for all their channels, and they covered the 2012 Olympics with rewindable live high-def feeds for every sport.

I found this really interesting. I would love to know more about the business relationship between the MLB, the teams, and BAM. On the surface, this reads like the MLB simply took some huge bets that paid off: 1) $120MM in funding, 2) control over ticketing, and 3) perhaps most shockingly, not giving up on the team despite some expensive, early failures.

I imagine there was quite a bit more adversity than the story hints, though.

Good for the MLB. I'm not used to reading about high-value tech companies spinning out of incumbents.

> BAM doesn’t run like a startup — it doesn’t have equity to share with employees, causing it to lose out on talented engineers and developers that want upside beyond a salary.

Is this really true? I imagine a lot of people fit this description, but just as many (talented) people are looking for a stable, well-paying job with interesting challenges (which this definitely seems to provide).

Unfortunately the minimum hardest requirements for PC mean that although my little laptop streams Netflix, Amazon and YouTube just fine it stutters on HBO Now.

It's too bad because I feel like they're missing out on a significant portion of their market.

Can anyone comment on the technical architecture this kind of system uses? I assume data flows are TCP instead of UDP? As soon as you let anyone pause or skip you need to be streaming different data to each client.

It's an excellent platform. They also developed the WWE Network, which has been fairly rock solid since it went on air last year.

Another nail in cable TV's coffin

Reply on news.ycombinator.com