Hacker News Reader

Comment by rocky1138 | Hacker News Reader

> This shows "a new Sony" and should be applauded by investors, SMBC Nikko Securities Inc. analyst Ryosuke Katsura wrote in a report. “Management is adapting rapidly to change."

Uh, no. This shows that the developers have absolutely no control, say, or ownership over their work. Things like this happen in companies (or divisions within a company) right before their best people jump ship because they are so poorly respected they are literally not given the time of day by management.

I'm reminded of Ray Kassar's comment about programmers being at the same level of someone who puts the game boxes together.

View on HN

I'm a developer and low-level manager, and wouldn't expect to have much input in this kind of decision. Especially in a behemoth like Sony. This is C-level-executive decision making territory.

I might have something to say about the idea from a technical perspective, if asked, and if that aspect of the platform were my purview, and I suspect they discussed it with plenty of knowledgeable people at some level.

I'm sure some people who were developing their in-house cloud solution feel like they've been blindsided, but if they'd succeeded, they wouldn't be in this position. The article says they've been struggling with it. They likely had a chance and missed benchmarks one time too many.

They're likely better off using a major cloud provider anyway. They've already solved the big problems.

There’s also the fact that if you release this kind of consideration to rank and file employees, you take the risk of someone leaking it to the media or insider trading on it

I have to wonder if any of these cloud gaming initiatives will succeed. It's been fun watching people throw money into that pit.

The problem seems to be people are unlikely to pay large subscription fees, but they also generally play at the same time of day so you can’t get a large multiple between hardware and subscriptions. Worse, people still need hardware at home to connect to these services.

~130x as many people bought PlayStation 4 than subscribe to PlayStation now at just 20$ per month. While they might be extremely profitable for Sony, AAA games require huge customer bases.

That's a general problem with all-you-can-eat subscriptions whether games, Adobe products, or O'Reilly books. If you're a fairly heavy and regular user, subscriptions are usually not a bad deal. You get access to try new things out and you have a known predictable cost.

However, most people are relatively light and irregular users. For a $5 a month service, that may be OK. But as the price goes up it gets less and less attractive for anyone who isn't a heavy user.

I could get on board with this if the devs had a say in setting those benchmarks.

Any good executive consults with their subordinates to create realistic objectives. This is so fundamental that it's implied. That's the CTO's job. They will in turn rely on their subordinates all the way down to the lowly developers.

Are you suggesting that developers deserve some kind of vote? They're a cog in the machine. Sony isn't a democracy or a worker's collective. As a developer myself, I'm not trying to diminish the role. I like to think I can provide a valuable perspective and that I'd be consulted about my small part. I'm certain that's exactly what Sony did, because they are a successful, well-run company that's been doing this longer than I've been alive.

Beyond that, no, the developers shouldn't have much say. Their role is to help determine what's required to achieve the objectives set by management. Good management will consult with them, set goals, and change tack if the objectives aren't met. If they still can't achieve the objectives in a timely and cost-effective way, management will replace them and go with a vendor. This is all right and good.

Are you suggesting that developers deserve some kind of vote?

Of course not.

Their role is to help determine what's required to achieve the objectives set by management. Good management will consult with them, set goals...

This. Consider yourself lucky that you've never worked someplace that starts off with "These goals are aggressive, but in my mind achievable" and then proceeds to give you objectives that are a wildly optimistic extrapolation of a of a previously achieved timetable.

This isn't just developers, either, but that's the context here.

Any good executive consults with their subordinates to create realistic objectives. This is so fundamental that it's implied.

...

I'm certain that's exactly what Sony did, because they are a successful, well-run company that's been doing this longer than I've been alive.

If you're right them I'm onboard.

Would they set reasonable benchmarks, like market share and revenue? That is what matters in strategy decisions.

Why would they set those benchmarks? Dev benchmarks are things like delivery schedules, development budget and capability trade-offs.

It honestly shocks me the degree to which some software developers are able to live with a complete lack of self-awareness. Developers absolutely rant about product managers giving any sort of technical feedback, and yet they're perfectly willing to act as if they should be included in corporate level business strategy.

I don't mean all software developers, but it's a common refrain on this forum.

The thing about developers, and "tech" people in general, is that they're prone to believing they're smarter than everyone else. I'm not sure why that misguided notion persists in our culture, but it does.

Convince me otherwise. There is a reason why developers make how much they do despite most developers not having the social skills to negotiate aggressively for good comp.

> There is a reason why developers make how much they do...

The reason is that they're temporarily in demand. Ask the people who were around during the dot-com bust how much the average programmer was valued then.

> This shows that the developers have absolutely no control, say, or ownership over their work.

Developer here. What role would you see developers playing? This is more strategic than technical, so I wouldn't expect much developer input at this stage.

If management asked and the developers didn't like it, people would say "why the opinion charade?". So the only options are don't partner or get a strong (whatever that means) majority vote from the developers before doing anything. Both seem like bad options to me.

Let me try (it's a long one)

Software is a new form of literacy. Literacy makes all the difference - you don't hire illiterates to run companies, they might be clever, politically astute, but the decisions they make will be wrong from the important perspective of "does this make sense in a literate world".

Imagine you were hiring people for a trading company in the 16th Century. If the top people could not read or write, would you expect that trading company to perform as well as a competitor that had hired its board, its execs, as those who were able to red and write. Who expected to read the ships manifest?

Companies that ignore the developers are companies that are ignoring what literate people think is sensible. Now sometimes that will be the "right" thing to do, but increasingly *when illiterates and literates clash over policy, the literate side is more likely to be the right one to take"

Just my 2 cents

It's just about different job roles.

Do you consult with the admin staff to check they're okay with your code?

In large businesses (in particular) you have to have delineation of roles - everything can't be democratic. Just as (I assume) you expect your seniors to trust you to code competently, so you should trust other people to handle strategy competently. You might not like the end choice… but that’s usually because via not having been in the discussion, you lack key information as to why the other smart people in your company came to that end choice.

(One thing I have learnt: everybody wants to be involved in strategy, because it’s fun, and because they think they can. I see this in my own industry all the time. People should have the discipline to stick to the scope of their role.)

Late reply I know but ... just replace the words "company strategy" in your last paragraph with "everybody wants to be involved with "electing parliament" ... stick to the scope of their role.

One of the things that came with increased literacy was increased middle class and increasing widening of the suffrage.

I honestly don't understand the point you're trying to make. As far as I can tell you are saying that people that can code are fundamentally superior to everyone else, to the point where everyone else is useless not only to corporations but society?

It's helpful to understand the fundamentals of the area you're making strategic decisions about. Even better if you're the one who has to deal with the aftermath, so that you have some incentive to learn for the next time.

The typical C-level dictatorship provides neither of those (besides just being a plain miserable environment to work in).

I think the C-Level dictatorship know the relative strength & weaknesses of their company. Sony isn't good at online services and its infrastructure relative to other companies like Microsoft and Google. It took them a decade to allow users to actually change their user names on PSN because they were effectively using the username as a unique primary key and associating all game save data against locally and remotely.

How are they going to make a Google Stadia style service that requires custom hardware units in data centers around the world with the lowest possible latency to the end user? And how are they going to do it quickly?

I guarantee Microsoft has a competitor to stadia ready to go, they have the data centers, they have the development expertise and they've been pushing xbox apps onto every platform that will take them.

If Microsoft will give them a good long term rate on the custom azure instances that the xbox team will be using to host the games then that's probably a whole lot better than Sony trying to roll their own.

>>How are they going to make a Google Stadia style service that requires custom hardware units in data centers around the world with the lowest possible latency to the end user? And how are they going to do it quickly?

I mean....they already do, so in that sense they are ahead of Stadia. Playstation Now is a real service that exists and which you can use already, and which required custom hardware in datacentres around the world. Now we can argue whether it's as good as Stadia promises to be, but that's purely theoretical at this point - one product is an actual commercial thing you can get, the other one exists in beta stages.

Just let's assume the basic tenet is correct - software is a new form of literacy [#]. No I do not believe that people who are functionally or totally illiterate today are useless or should be discriminated against. But the value that an illiterate person in normal circumstances can deliver is dramatically reduced compared to the same person but literate.

And I am simply transferring the same intuition about this to software.

If "Google SRE is what you get when you ask a developer to design an ops team" then what do you get when you ask a developer to design a whole corporation. Or a government ministry?

People who can code are not fundamentally superior to everyone else as you say, but they offer different ways to organise as much as people who 300 years ago were literate were not fundamentally superior to the rest of us, but offered different ways to organise.

So here is my simplest idea - as pretty much everything is being eaten by software, then project management should flow entirely from code. How is Project X doing? Ask the codebase.

[#] Bicycles of the mind if you wish. But something important is happening - its beyond telegraphs or telephones in enabling communication between humans.

Literacy is the ability to read and write regardless of the domain (science, history, poetry, religion, software, whatever).

Software is, ultimately, a set of logical instructions. Other people who understand those instructions are literate in that domain. You seem to be taking a single application of literacy (software) to be the entire scope of literacy writ large. It isn't clear to me why you think that, especially when no programming language is capable of "encoding" the range of human thought that, say, English is.

Please respond in the programming language of your choice.

You seem to be saying that software is one "silo" of literacy, like the jargon one needs to understand film criticism or the politics of the hundred years' war.

I don't think that's fair. Software if anything is a means of modelling and transmission of that model - being able to create and manipulate models is what we try to do with language and written word - but the compiler we are targeting is on other people's heads.

so it's a bit hit and miss - we spend a lot of effort making us all share the same compilers, from schools to legal judgements.

but with software the compiler is simpler and so the models can be more transparent and that Inthinknis the big difference.

No I cannot express my model in english as well as I can in any program- I am not trying to express the whole range of human thought in software - but what we can do is and will be so valuable I can only compare it to the shift in value we got from mass literacy

Not jargon. The ability to read/write code is a knowledge domain. Literacy allows communication about all sorts of knowledge domains (software being one).

But knowledge of the software world does not (dis)qualify one for business leadership, or to be a heart surgeon, or a janitor, or any number of fields. Your original post implied that knowledge of software, the "new literacy" is a prerequisite for any moderately advanced task, which is nonsense.

I think your point is that software is like "engineering" "Biology" or "house building" - one siloed skill amoung many that some people will specialise in

I disagree - I think software is more horizontal - it allows great improvement of any form of knowledge work, and is much more comparable to "learning to read" (or perhaps sits somewhere in a hierarchy of horizontal skills between "reading" and "statistics")

Does it really mean anything?

The PS4 and Xbox One use very similar hardware. In fact, other than DRM, neither is that different from a gaming PC.

It seems you'd have no trouble running PS4, PS5, Xbox One or Xbox Two games on the same hardware box that would be racked in an Azure data center.

For that matter, Microsoft has Azure data centers all over the place and very good networking for those data centers. A few years back I noticed that the nearest Azure data center had 35 ms latency to me compared to AMZN's us-east-1 which had 50 ms. (Maybe that is why AMZN built us-east-2 across the street!)

I don't see any reason why Sony should build a whole bunch of data centers and struggle with second-best networking when MSFT has already made the investment and is continuing to make the investment in a great data center product.

>I'm reminded of Ray Kassar's comment about programmers being at the same level of someone who puts the game boxes together.

As someone who has worked in a number of places and ended up programming... I find this absurd.

Lots of programers have no idea how good they have it / influence they have.

Just because someone could make a call without their input doesn't mean they are like manual labor.

There's no evidence that no technical consultation was done. It's very likely that senior technical staff were consulted, and asked to keep it quiet (and succeeded).

All this report says is that rank and file had to be calmed down after the reveal (fair enough). That doesn't mean that technical input wasn't asked or, or even listened to.

Like, I can totally see something like HQ asking "What would it take for you to grow X, achieve Y, by Z", and then PlayStation unit saying something like "M people, N money, O risk". And then HQ was like "oh, MS can do better than that".

Lol yeah the developers build the thing and the business people run the business

Tech companies constantly have teams building tools and products that never ship because strategy changed. The developer-centric FAANGs probably do even more of this than the more traditional big corporations.

Reply on news.ycombinator.com