Hacker News new | past | comments | ask | show | jobs | submit login
No Copyright Law: The Real Reason for Germany’s Industrial Expansion? (2010) (spiegel.de)
429 points by gballan on June 16, 2018 | hide | past | web | favorite | 168 comments



I had no idea about the situation in Germany, but it's a well-known fact that the current power of the Swiss chemical and pharmaceutical industry comes from the lack of patents law in Switzerland during the late 19th century[1]: a few French chemists moved to Switzerland to circumvent the French patents.

A well-known example is Alexander Clavel who, in 1859, founded Ciba AG, a silk-dyeing business using a French patent with no license. Ciba later became Ciba-Geigy, which later became Novartis. One of the biggest pharma corp in the world today.

Source: Swiss National Museum, Zurich.

[1] : Patents were only introduced in the Swiss law in 1907.


It is likely that the lax Chinese attitudes towards patents is one of the contributing factors to their highly agile and efficient technology sector.

It has been noted for some time that one of the barriers to entry for a new company in the US is their lack of defensive patents.


The part of the industrial revolution that happened in the US was similar (maybe more trade-secrets related):

> In the UK, he was called "Slater the Traitor"[2] because he brought British textile technology to America, modifying it for United States use. He memorized the designs of textile factory machinery as an apprentice to a pioneer in the British industry before migrating to the United States at the age of 21.

https://en.wikipedia.org/wiki/Samuel_Slater


And it wasn't just patents. The US was notorious for not respecting UK copyrights for most of the 19th century.

https://en.wikipedia.org/wiki/The_Pirates_of_Penzance#Backgr...


"defensive patents" Yes, makes it sound like patents are a legalized protection racket.


Right, protecting an investment in R&D is a protection racket.

There has to be some reasonable line between incentivizing investment and "everyone had a right to everyone else's ideas".


If it is necessary with defensive patents for a new company to enter the market with an obviously not - in a quite literal sense - copied product, then it probably is something comparable to a racket.

If it is also the case that patent holders purposely withold the very trade secrets that is the bargaining chips that gets you a patent according to the spirit of the law, then if it isn't a racket, it's still immoral and needs to stop being allowed.

So many patents these days are completely devoid of the secrets they are supposed to both protect, and publish. Little is thus gained for the public good, and much is lost.

It strikes me as odd that patents has such a strong support in the US, as generally there seems to be the case that the government isn't wanted where it's not absolutely needed, and I find it hard to see how one could argue that patents are so integral to society that they should fall in that category according to the metrics that seems to be used for other commonly government controlled areas.


The article argues otherwise and backs it up with a plausible story from history, with two other stories (Switzerland and China) added above in the comments.

At least it seems plausible that the current way of "incentivizing investment" is not working, and we might even be better off completely without it.


>Right, protecting an investment in R&D is a protection racket.

It is, when most "R&D" is trivial compared to the revenues you command, and is stealing off earlier ideas, government subsidies, and so on, anyway...

Besides, you said it yourself: "protecting". It obviously is a protection racket, and is an artificial limitation imposed on the market by legal dictum.

Artificial because, unlike e.g. police protection against mere theft or contract law, this one, as history shows, is not necessary for the market to operate and grow.


Agreed for more involved research, but for software, patents are a scam.


it is often said that software is just math, and math is not patentable.

But any mechanical contraption is ultimately just physics at work, and physics is just math as well. So, what's the difference?


There is nothing about software that makes patenting it inherently any worse than patenting hardware. The problem is rampant misuse/abuse/poor implementation.


Maybe, but in the software sphere you have the situation where the most obvious solution to a host of problems have been patented by various companies and it's not feasible to do a patent search on every algorithm you implement. Especially since many of the patents are quite broad and cover entire fields of technology.

This is also why patent trolls are such a nuisance, since defensive patents don't work on entities that don't produce anything.


>There has to be some reasonable line between incentivizing investment and "everyone had a right to everyone else's ideas".

Isn't this just argument to moderation?

Moreover, patent portfolios that inhibit competition through the threat of a lawsuit are disincentivizing investment and competition.


> There has to be some reasonable line between incentivizing investment and "everyone had a right to everyone else's ideas".

You're meant to pick two extreme positions and imply the reasonable compromise is midway between them.

Instead you chose one extreme and one reasonable sounding position.


Honestly not sure if you mean the first or second is the reasonable one.


Speaking for myself, everyone does have a right to everyone else's ideas, even in patent systems. Patents are supposed to be a temporary monopoly to encourage investment, but quite a lot of investment happens anyway even when patents are not bothered with.

Not all, certainly, and I cannot quantify how much as a fraction. However, I can give the anecdote of a former friend who worked in audio needing to get a patent on something (encryption I think) to get industry certification for their products. Their patent was modified by the lawyers from something novel to something which already existed, but the patent was granted anyway and nobody really cared because the patent was only applied for to get the certification.


Form a startup that offers legal protection against patent trolls for $1-2,000/mo and you will have a winner...


There are patent groups you can buy into (with costs being less if you have helpful patents to share) and then use a shared patent portfolio to defend yourself.

It isn't a startup because it needs a lot of resources and the patent portfolio isn't very useful against patent trolls because they intentionally have no production that could be infringing on anything. But I think they also act as insurance for defense costs.


Patent groups are not the same as Patent Troll insurance imo. Latter would be much better (even if you would need to give up all patent right).


>Patents were only introduced in the Swiss law in 1907.

Luckily for us all otherwise we might never had a certain patent clerk figuring out relativity.

Edit: Also the Swiss Patent Office was founded in 1888 which is more inline with what I remember as Einstein worked there since 1902.

Do you know under which mandate did it operate then?

https://en.m.wikipedia.org/wiki/Swiss_Federal_Institute_of_I...


https://ir.lawnet.fordham.edu/cgi/viewcontent.cgi?article=14...

Germany and the United States put (tarriffs) pressure on Switzerland and the turn of the century, forcing them to protect their patents in Switzerland. They unhappily obliged after years of talks and trade issues - in 1907.

PS: specifically chemical patents


The then Federal Industrial Property Agency officially came into existence on 15 November 1888. This was as a result of the Patents Act of 29 June 1888 and the Industrial Designs and Models Act of 21 December 1888. These two new laws required a new federal office to assume the associated tasks.

https://www.ige.ch/en/about-us/the-history-of-the-ipi.html

Maybe what changed in 1907 was that French patents were protected as well?


Or the chemical process. But you're right, that was probably a simplification from the museum.


> Luckily for us all otherwise we might never had a certain patent clerk figuring out relativity.

Also glad he didn't patent it.


Only designs for machines and methods embodied in machines were generally patentable in the past. Mathematical formulae are not patentable.


The same is true of China, the US, and Germany. The idea that developing countries can get rich through free trade is a myth [1]. Developing countries that embraced free trade (see Africa, South America) largely remain poor today while countries that embraced protectionism (see Asia) are much wealthier. The basic fact here is that industrial societies need to be bootstrapped by a strong, active government (who can defend native industries and provide financing in the form of fiat currency) and rampant copying by those native industries (who have no legacy investments and can jump straight to best-of-breed practices and the latest technologies). None of this is a new idea: it was all laid out by Friedrich List almost 200 years ago. Unfortunately the basic idea, despite the mountains of empirical evidence, fell to the onslaught of "free trade" propaganda.

What's interesting is that the same basic dynamic likely applies to even highly developed economics. The tariffs are likely just the beginning. We will see even Western countries abandon "free trade" and strong, active states taking firm control of the economy in order to drive growth. The key words that will define the economic future will be "political economy."

[1] https://www.nytimes.com/2018/02/07/magazine/the-rise-of-chin...


>Developing countries that embraced free trade (see Africa, South America) largely remain poor today while countries that embraced protectionism (see Asia) are much wealthier

The assumption that South America and Africa are pro free trade is not true. Indeed, most of the countries with the highest tariffs are in Africa.

https://www.tutor2u.net/economics/blog/which-countries-have-...


I don't think it's right to say that enforcing copyrights is an aspect of free trade. In practice many countries don't want free trade unless their copyrights get enforced and so copyright ends up getting rolled into free trade agreements but you could say the same with health and safety standards in products.

I also think of government support for industry as orthogonal to free trade. Yes, countries have certainly sped up their development through the use of tariffs. By transferring money from consumers to people accumulating capital they increased the effective savings rate of the country and increase growth. But you can also literally just tax consumers and pass the money on to industrialists in the form of cheap loans like the Japanese did in their development. They didn't go off the gold standard until the Great Depression but I'm sure you could industrialize through fiat currency too though no cases come immediately to mind.

I'm not entirely happy telling current poor countries "Suffer now so your grandchildren can be rich" even it that's what my ancestors did for me. And I'm certainly not going to advocate the sort of violent labor movement suppression that seems to have always gone hand in hand with rapid industrialization[1]. I suspect that a lot of truths about the world are going to change as AI develops over the next half century so I feel like I shouldn't be advocating sacrifices I'm not sure will pay off.

[1]https://pseudoerasmus.com/2017/10/02/ijd/


Your point about developing economies is true, but this:

>What's interesting is that the same basic dynamic likely applies to even highly developed economics.

is not


"political economy"

Marx used this a lot 200 years ago.


I’ve got no clue what this has to do here, but all that article shows is that The NY Times is doing a swell job in providing substance free material and a love for the almighty state.

The author of that article is this fella [0]. While I do not doubt his writing abilities, I do doubt his expertise in economics and his political motivations.

Simply put, I do not show up at X field and pretend to know things just because I think I’m smart, even if I don’t have anything to do with said field. It’s interesting how with economics everyone thinks they can do just as much.

If you would like a basic intro to free markets checkout Milton Friedman’s Free To Choose series on YouTube. It’s from the 80s but well worth the time investment. It’s an easy to understand documentary series and at the end he talks to a group of people about the subject shown.

[0] https://en.m.wikipedia.org/wiki/Pankaj_Mishra


Same thing happened in the Netherlands. During a six-year period where they didn't recognize international patents, the country was able to rapidly catch up in industrialization. The technology company Philips owes its success to being able to copy British lightbulb designs during that period as well.


Wasn't it the same for the Swiss watch industry? I think they basically copied and then perfected British designs. To expand on this, I'd say Switzerland and Japan have remarkably similar early paths to wealth.


They also had fine metal workers who had a much smaller jewelry market with Protestantism.


That’s where the ‘perfected’ part comes in. I wonder what they did before watches or how Western Switzerland got good at metalworking from nothing.


For all the "intellectual property" insanity that comes out of the movie industry, one contributing factor to the rise of hollywood was the difficulty that inventors had enforcing their patents in California.


On the other hand, it is believed that British patent law is one of the reasons most of the great inventions of the industrial revolution happened in Britain. James Watt ultimately died a very wealthy man because if it [1] and it encouraged others to get wealthy through invention

[1] https://en.m.wikipedia.org/wiki/James_Watt#Patent_trials


Patents therefore seem to be excellent at producing very wealthy men, which is not necessarily the highest social goal. It's probably good to try to capture some of the social value of a "new" idea for the originator. The specific mechanism of patents is a huge drag on human industry which may not be worthwhile.


Same was true with Italian pharmaceutical industry until about 1960.


Curiously, we see a similar situation today with open source software. The free nature of it would seem to stifle creation of top quality open source software, but the reverse seems to be true. The open source business is thriving.


I think a more pertinent example is the Shanzhai culture in Chinese manufacturing, which is distinct from both the overtly open culture of Free software and the explicitly closed structure of intellectual property.

You're unlikely to find a datasheet for the latest MediaTek chipset through Google or Baidu, but you won't get sued for sharing a copy with a friend. You can't buy sample quantities through a distributor and MediaTek won't talk to you unless you're a serious player, but it's not hard to buy a couple of chips or a few reels if you know the right person.

Information isn't free, but it doesn't have a price tag either - it's exchanged as part of a reciprocal guanxi network. Much like the BBS warez scene, information gradually percolates from a core of well-connected peers to a wider network, creating a hierarchy of esteem and a duty of reciprocity that incentivises the continued flow of information. I share with you today, in the expectation that you will share with me tomorrow.


In my limited understanding, guanxi has the taste of corruption, not openness. People doing favors for those they know rather than a free, open, arms-length market which provides more liquidity and opportunity. Is that wrong?


Yes. That's how people like Elizabeth Holmes raise a billion dollars and gain a who's who board comprising of famous statesmen without any questions asked, how young son in laws become more powerful than career diplomats and how C- suites always get other C-suite jobs.

At least this Chinese version in the electronics is more democratic and let's the little Chinese startup try out things quickly and get into a full commercial relationship if things succeed. I understand American companies used to do this early on but eventually rent extraction mindset took over.

But this behavior extending up into actual corruption is very much a current problem for that country


Did you think that someone was criticizing Chinese culture and it needed defending? And the defense is, there is some degree of corruption based on relationships in other places? Every place sometimes has bad weather too; does that mean the weather is the same everywhere? And does the fact that there is also bad weather in Rome have any bearing on the weather in Edmonton or help the people there?

Addressing guanxi seriously: My understanding is that guanxi works against opportunity and the free market; not anyone can get a loan at the bank, for example; instead it requires personal relationships. Economists I've read said the lack of arms-length commercial relationships hold back economic development and opportunity in China, but that was written maybe 20 years ago and possibly things have changed. I do know that Communist Party connections, and particularly connections to the 'royalty', the families of leaders, are highly valuable and are at least sometimes seen as necessary for success in business.


Guanxi is a broad term and can mean anything from networking (the business kind) to asking about job referrals from friends, you can google the investopedia definition to get a better idea. It can be seen negatively or positively based on the context as in like "he got his job through guanxi" aka he either networked hard (positive) or there was some kind of nespotism involved (negative).


That seems pretty close to how westerners use the term, unless I am missing something?

Context matters, I think is what we can all agree on. Unfortunately, in order to agree on the appropriate context, we have to see each other as peers and working for the same goal. (I think this is what's fucked with the current American political system -- it's less "you have your facts, I have mine" and more "you think this context/lens is what matters, I disagree and choose these base assumptions."

Naive realism wrote bigly.


Yep, guanxi is roughly the Chinese word for "connections" or "through the network".

I think its perceived more negatively nowadays by the Chinese with 'newish' (more 1990-2000s) terminologies such as the "关系户" (guanxi-hu) which is roughly translated into English as person with a lot of connections and/or is politically savvy and can gain a large advantage in society or conducting business through their network. Note that Chinese govt. (especially the lower rungs) tend to be corrupt/non-transparent and society is more about giving gifts (originally between friends and families, basically bribery if asking for a favour) so this tends to give a much more unfair advantage than Western society (the politically transparent ones).

Chinese people (usually media) either think negatively of these kind of people as they see them having an unfair advantage while some respect and aspire to be guanxi-hu because of how well-connected they are.


People doing favors for those they know rather than a free, open, arms-length market

This describes most markets that are not in fungible commodities.


It's not really "favors". The person selling individual chips profits because they don't have the overhead of a formal business. There are open markets for formal businesses, which also expect you to be a formal business ready to purchase large volumes.

and, tbh, transparency doesn't matter at low volumes. If someone can afford to carry a prototype to production, the difference between the average and lowest potential guanxi price for a single chip is negligible.


I like this example better too, or at least I think it's interesting.

I think one key is that Shenzhen's manufacturing culture was not too centralised or formalised. This let a culture to emerge. I think your emphasis on is right. Culture, not ruleset or ideology. Emergent, not designed by some philosopher.

Most people want an absolute answer, an absolute truth about intellectual property. A clear right and wrong that tells us what IP laws should be. The answer needs be true for chemistry in the 18th century, aeronautical engineering in the 19th, medicines in the 20th and software in the 21st. I don't think this truth exists.

Ideology is not the way out of this. Ideological thinking (eg Stallman, Rand, Marx, Plato) is important as an exercise, but when people actually buy into ideology we're doomed to fail. The way an ideological thought process goes is usually:

(1) Start from fundamental truths, usually moral truths. Property is exploitation. Unfree software is immoral. Selfishness is the only true love. Platonic forms are the truest truths. etc. Prove your truths metaphysically. Hold them self evident. Make them immutable.

(2) Find some complicated but (seemingly) airtight way that everything good about life, the universe and everything will inevitably follow when people accept the Fundamental Truth. Software will be better. The economy will be better. We will achieve a classless harmonious society. Technology will advance. etc.

To do this properly, you'll probably need to produce a very long, very boring book that works out all the details. This is how roads and police will work without government. This is how the economy will work without money. This is how software will be an industry when everything is strictly GNU. This is how hover boards will be invented.

This step is inside baseball. Anyone who can slog through "Objectivist Epistemology" or "Das Capital" is... you need to be a hardcore believer.

(3) Keep adding to no. 2. The more pages, the better. No. 1 can never change. No. 2 now needs to explain the entire world. Don't give up. Keep going. There is no such thing as a corner you can't argue your way out of with enough new terms. Degenerate Socialist State. Core-periphery modes of production.

Keep going. Stay motivated. There is no such thing as a corner you can't argue out of with enough focus on a relevant totality, the transitory nature of social forms and a dialectical sense of motion through conflict and contradiction.

...This road goes nowhere. There is no absolute truth of life the universe and everything. You asked the wrong question.

I totally agree that the IP/copyright systems that exist today and are being promoted as the no. 1 priority in international law... they are not good systems. At least, it would be very helpful to have other systems as well. The best systems will probably emerge, not be designed by lawyers.


> The open source business is thriving.

Thriving, may be, but not exactly dominating the consumer market. The vast majority of software companies still keep their code well under wraps.


In the article it mentions the once popular book 'Principles of Leather Tanning'. This probably sold really well but not as well as shoes, belts or other finished leather goods. People would probably have at least double the amount of shoes as copies of the book. The book was for the trade audience not the end consumer.

Open source is a bit like that. Much like a book has to be read rather than merely used/worn, Open Source Software has to be cloned/forked/apt-get-ed/'make'd and so forth.

Now if you were to buy a pair of leather shoes then the particulars of what chemicals and processes were required to tan the leather may well be learned from the 'Principles of Leather Tanning' but tweaked a bit to a proprietary recipe that could be kept 'under wraps'.

Regarding the Open Source desktop applications, Libre Office is not a clone of Word/Excel/Access/Powerpoint. Same with Gimp, it is not a feeble clone of Photoshop. These tools may be able to use the same file formats but there is a different ethos to them. For some people the UX of Libre Office is far superior to MS Office - no ribbon and therefore no need to Google search the simplest of tasks. If you are dealing with data - as in CSV files - then OpenOffice is your friend, MS Office just ruins it. This does not affect people doing glorified to-do lists and glorified receipts in Excel but if you are reading in data from some code you have written then OpenOffice is the more productive tool.


depends on what segment you are looking at (phones anyone?).

Never mind that MS have a long long history of playing dirty.

I distinctly recall that back when the netbook first appeared, MS did a 180 on their efforts to kill Windows XP because Vista could not run on such low spec hardware.

Never mind that later on you could find Dell selling Windows equipped netbooks with a rebate campaign that made the higher spec Windows variant actually cheaper than the lower spec Linux (Ubuntu) variant (what is the chance that MS was funding that campaign?).

Damn it, these days Microsoft waves the OEM license fee for tablets with a screen below 9 inches.


Malware shipped by laptop manufacturers on Windows laptops funds Windows being cheaper than Linux.


I will grant you most laptops don't run an open source os, but most of the times they are probably running a browser -- and both firefox and chrome are almost open source.

For mobiles it is a little different, because iOS locks people to their browser, but at least on Android users would be using Chrome.

Heck I am typing this on a chromebook, simply because I realized 80% of what I do is webbrowsing anyway.


The majority of servers are open source, the majority of consumer devices are open source, all super computers are open source. How is open source not eating the world?


As long as MS Windows still has 89% market share on the desktop/laptop market [1], as long as Google still keeps core parts of Android under lock and key, and as long as all my friends still communicate with Facebook and WhatsApp, I think the claim that "the majority of consumer devices are open source" is a bit early.

Look, guys, I'm not saying open-source isn't important! And I'm definitely not saying there aren't any great open-source programs out there! All I'm saying is: on the current market, most companies that produce software choose to distribute this software under a closed-source license. What's so hard to accept about this simple observation?!

[1] https://www.netmarketshare.com/operating-system-market-share


Open source makes sense for widely used, preferably infrastructural things. Operating systems, various network servers, programming languages.

Most companies' has a very limited audience and a very narrow purpose; it can't benefit from the network effect.


You just named 0.1% of the revenue of the tech industry.

Closed:

Google search. Windows. Office. Facebook. Cisco. Intel. Amazon AWS. Amazon retail. Most Adobe products. Most Oracle products. iOS and most of what Apple does. Salesforce. Workday. Most of what nVidia does. Most of what Qualcomm, Texas Instruments, and pretty much every other chip maker does. Most of what Samsung does. The core software of: PayPal, Square, Stripe, Visa, Mastercard, et al. The list keeps going.

Most of what every major tech company does is closed off.

The tools are largely open source. The actual pieces of the companies that generate revenue are not. They love open source, when it isn't about their revenue generation.

Ask Facebook to show us all of their source code.

Nearly the entire tech industry is built on massive, closed products.

Google search by itself generates more operating income every year than every public open source company combined, by a factor of ten. Ask Google to show you all the source code to search.


That we no longer pay rents to the owners of the kernel is a huge victory for everyone. That you mistake market failures for successes is a rather sad reflection on the state of the soul of most technologists.

The commodification of services that used to be cutting edge happens all the time, and if not for artificial patents would happen even faster.


Pay rents ? Do you think the Linux kernel just magically appears out of thin air ? If not, then someone is paying for the kernel development. If it isn't you, then that organization(s) or person(s) may have goals that are completely antithetical to your goals.


Or I develop part of the kernel for something I need for my day job, send it up stream to make my life easier the next time I to use it and someone accepts it for the same reason.

You then need someone to pay what is essentially an administrator to keep track of the changes.

The reason why a company would do that is because of the name recognition advantage they get from this. "Oh you need someone to tune your Linux performance for your super computer cluster, we have the guy who it's named after on staff."


Yes, that does indeed happen, and is an example of where you're paying with your time.

My point is that if you're not paying with your time, and most people are not, then someone else is paying for it. The rent doesn't go away, it's just getting covered by someone else.


It's paid for by all the entities paying Linux developer's salaries as well as everyone else that contributes with their free time. Those companies probably build that cost into the price they take for their products, just as they would do with licensing costs for proprietary software.

But it's not a wash. The difference is that with Linux all those contributors not only get a usage license for an OS but they also get all the control over it.

You're a startup with some new kind of device or service? Good luck getting $PROPRIETARY_SOFTWARE ported to it if you're not HP/Dell/etc. and can guarantee sales in the millions. Heck, even HP didn't manage to convince Oracle and Microsoft to continue their IA64 support, despite major dollars being involved.

Where I work, one of two reasons for saying "we can't do that" is when hitting the limitations of nature (obviously). The only other one is when proprietary software is involved, commonly encountered with e.g. networking gear or server/BMC firmware.

While we don't get to not obey the laws of nature, the other one just sucks. It should be us doing the tradeoff between the benefits of doing something and the amount of work required to make it happen. It shouldn't be up to some 3rd-party vendor which is unlikely to consider us important enough to quickly implement a feature we want or heck, even fix a bug without having to wait $LONGTIME. With free software, that's exactly how it is.

If you want control over your own fortune, you don't get that with proprietary software.


1) That's a pretty narrow use case that doesn't apply to most users or developers, so, while I'm sure it's a problem for you, I don't think it's an example that applies to my original statement.

2) There's nothing about commercial software that says that you can't give out/sell the source code to all or some customers.


It would be a lot more than 0.1% if infrastructure software was closed source. Imagine if you had to pay Oracle $100k per server for your database.


supercomputers are not open source. They may use a linux kernel but they need a proprietary compiler and a bunch of proprietary drivers to interface that are custom made. pretty bad example to pick.


You’re asserting that so long as any code is closed, the platform is not “open source”. I disagree. When 99% of software on a device is open source, open source has clearly eaten than device.


well for supercomputers you are really far from that kind of ratio. even desktop computers are far from 99 percent for that matter.


I bet you’re not that far from 99% on supercomputers running Linux. Drivers are not the bulk of the system.

Obviously desktops aren’t at 99% unless you’re running Linux.


> I bet you’re not that far from 99% on supercomputers running Linux. Drivers are not the bulk of the system.

No, compilers should also count, and i'm not just talking about drivers. Everything from data storage (file system) to data transfer between the different racks of a supercomputer is usually proprietary. Seriously, look it up.

> Obviously desktops aren’t at 99% unless you’re running Linux.

No, even your Linux desktop is far from being 100% open source. How about the BIOS? The Intel microcode? Even open source GPU drivers interface with proprietary interfaces on the actual hardware. And let's not forget that many Linux distros don't make it very clear if what you are installing is actually free software or a proprietary blob. The reality is that it's very hard to have something fully running on free/open source software from A to Z.


I’ve worked on clustered computers. It’s been a while, but I doubt they’ve progressed to more closed.

I feel like you’re listing small pieces of the system and for some reason acting as if those are the majority of the code.


The relevant portions of a supercomputer are not open source: The MPI implementation, the compilers, the CUDA drivers, some of the tools to monitor the system, the fancy 3D AR wall software and so on...


Maybe the community, the "business" is doing quite poorly.


Not really similar. Copyright applies by default and releasing software is voluntary.


> Curiously, we see a similar situation today with open source software

I strongly disagree. Here are four notable counter-examples:

1. Desktop Linux. I'm a religious Linux desktop user, but MacOS is definitely a better operating system.

2. Office suites. Libreoffice/OpenOffice/whatever-we're-supposed-to-call-it-these-days is stuck in the 1990s.

3. Videogames.

4. Large web services (Google Search, GMail, Facebook).

Notice that four these examples, where closed-source software outshines OSS in quality, cover 90+% of most people's computer time...

Open source software seems to do especially well mostly when the end-user of the software product is a programmer/software engineer/IT type.


How is LibreOffice being stuck in the 1990s a problem? MS Office was essentially feature-complete by Office 95, and LibreOffice does everything Office 95 does. It seems all the major changes since then (Clippy, Adaptive Menus, Ribbon Menus) have no value to experienced users.


LibreOffice is a pain if you want to do anything that looks remotely attractive. MS Office, unfortunately, wins hands down at that (and at general usability and power).

Being unable to use MS Office was one of my big regrets of working on Linux for a long time. When I finally learnt LaTeX, that at least removed the need for Word ;-) (Although I still miss PowerPoint, LO Impress really doesn't impress me very much at all.)


I find that remarkjs (https://github.com/gnab/remark) has removed a big class of situations where I previously needed PowerPoint.


And has the advantage that videos actually work reliably.


This is an important point and a really big advantage as I've painfully found out over the years when videos have not worked as intended.


How do you feel about the Beamer class for LaTeX? I suspect it's better than Impress (I strive to not actually prepare slideshows in life, but I don't always manage to).


I've dipped into it, but haven't had the opportunity to do much with it yet. My impression is that it's fantastic for preparing academic presentations (where it's all about the content and you want to be able to create some decent equations), but not necessarily what I would use to create a fancy product pitch or anything like that.


Beamer is nice if you have a few weeks on your hands to spare. :P If you need to throw together a presentation for 3 days from now and you don't already know exactly what it should look like I wouldn't recommend it.


I rarely have to do presentations and know little of latex but gave beamer a try recently when I had a presentation to do. It's very simple syntax and I learned nearly everything I needed in a couple of hours.

The last time I used powerpoint (several years ago) I found the learning curve much higher and even after I started getting the hang of things had to keep fighting it for stuff like code snippets.

Throw in a makefile with an inotifywait watcher and you've got a WYSIWYG interface.


Give WPS Office a go on Linux... it's been the most similar to MS Office I've found (in terms of both file format and UI).


I second this, Office has become less and less usable over the years. They keep hiding features and menu items that I use and constantly move around what is not hidden.

Libreoffice is a breath of fresh air for me to use.

The only thing they have improved in the newer office is the intruduction of 64 bit excel so I can do large (100x100+) matrix operations without external plugins.


Most of LibreOffice is pretty good, but Impress is awful. Last month I put together a presentation. I wanted three slides with a few bullet points and a handful of images, but I gave up on Impress after encountering multiple different show-stopping bugs.

The only good thing I can say about Impress is that every time it crashed, it successfully restored my editing state when I started it up again.

I don't think anyone cares about that beast enough to maintain it properly. It should just be scrapped and its resources directed towards Calc and Writer


> have no value to experienced users

That's precisely his point: Experienced users deal well with open-source software, because much of that was designed with them in mind. But in all too many cases, it doesn't play well with pretty much everybody else.


But office software is typically used in offices, and considering the enormous productivity gains from using it correctly, it makes sense for employers to either hire experienced users, or to train their inexperienced users. Tweaking the interface might help beginners feel more comfortable, but it won't make them work anywhere near as effectively as somebody who actually knows what they're doing.


> That's precisely his point: Experienced users deal well with open-source software, because much of that was designed with them in mind.

I'd call myself experienced but I could never use Open/LibreOffice. It wasn't just the UI; they just couldn't even open MS Office documents without destroying half of their layouts.


How much of that is down to Microsoft bribing people to support its own super complex, hard to implement standard OOXML in place of ODF?

https://www.pcworld.com/article/136627/article.html


I don't know.


When did you last try it? That certainly was the case 10 years ago or so, but current versions of Libre office at least in my experience, do pretty well with Microsoft Word layouts.

My other observation was that a lot of the "formatting" that I see in office documents is unnecessary and only there because the software makes it easy.


> My other observation was that a lot of the "formatting" that I see in office documents is unnecessary and only there because the software makes it easy.

The problem with that statement is: if a colleague sends me his Word document, I don't care how unnecessary his formatting may or may not be, I want to see the stupid thing on my laptop, preferably in the same state he saved it in! And right now, LibreOffice rarely delivers on that.


> When did you last try it? That certainly was the case 10 years ago or so

I tried it 5 minutes ago.

Here's an example: https://drive.google.com/uc?id=1YHaKdXh2olSeVdX8ybhF1SG6fyDm...


Is your link correct? Firefox on Linux tells me it's not redirecting correctly.


Yes it is (though a redirect loop wouldn't be in my control anyways). It works fine for me in both Chrome and Firefox on both Windows and WSL.

Most likely you're blocking cookies or some Google domains or something. Try on a clean browser. Or try

  wget --content-disposition "$link"


>MS Office was essentially feature-complete by Office 95,

Sorry but that comes across as a bit ignorant in my opinion. It seems to me you have never studied how non-programmer powerusers use excel/word/powerpoint. People are running their entire businesses on excel macros/ms access/and other cobbled-together systems. Heck, I don't think you could even do pivot tables till office 2007. That rules out a large chunk of the financial industry.


> MS Office was essentially feature-complete by Office 95

Were Excel data tables the same before 2007 (or was it 2010)? Right now they're very easy to use but not sure about earlier.


I'd argue they have negative value to experienced users.


I know plenty of professionals who work wonders with MS Office and who love working with it. Calling them all "unexperienced users" strikes me as pretty opinionated.

Just because I would rather solve the task with R instead of Excel doesn't make me a more experienced user, it simply makes me an experienced open-source user rather than an experienced Microsoft user.


What did Mac OS build off of? Would that be possible without OSS? Same for the large web services. How far would any of those companies have gotten without open source?

No one is arguing that unfunded projects will kick the ass of funded ones. However, innovation that can be shared will produce more innovation.


>Would that be possible without OSS?

In the case of Mac OS, probably. UNIX licenses are available and it's not like Apple has exactly turned Darwin into a vibrant open source community.

As far as large web and cloud services more broadly? It would certainly be a much different world.


Mac OS came from Mach, an open source microkernel from CMU in the 80s. Although Objective-C was originally proprietary, NeXT implemented it on GCC and open sourced it.

The origin of modern MacOS/iOS is definitely open source.


Absolutely. My point was that, absent open source alternatives, Mac OS [EDIT: OS X specifically] could have been created on top of a proprietary UNIX. The lack of open source Unix alternatives would likely not have had much of an impact on Mac OS and certainly Apple hasn't taken advantage of the open source development model for Mac OS.


> Mac OS came from Mach

I believe you mean OS X, not MacOS.

Apple could switch OS X to entirely proprietary underpinnings and few customers would even notice. In fact, OpenStep ran just fine on top of NT. [And also on top of then-non-open Solaris].

https://en.wikipedia.org/wiki/OpenStep


OS X = MacOS.

Apple confusingly renamed OS X to MacOS a few years ago.


obnoxiously, it's actually macOS, not MacOS.


Having been using Mac os for a while at work, I can't say it is superior. More supported, but equally as broken as my expectations for any os.

Office suites just don't build for Linux. Mainly because the people buying are not there, I would wager. For personal, I still loathe most.

Games, it's true. But the same situation. Luckily, I don't do 3d shooters anymore. Steam mostly works. Otherwise, most of my game time is on the PS4.


> Open source software seems to do especially well mostly when the end-user of the software product is a programmer/software engineer/IT type.

Yes. The article and the parent regard industrial benefits.

Most of the propriety alternatives that you mention derive a great deal of their value from networked capabilities, made possible due to the proliferation of GNU+Linux on servers.

In this way, the availability of GNU+Linux without traditional copyright restrictions has been a huge boon to the commercial software industry, just as the availablity of printed knowledge without traditional copyright restrictions was a huge boon to various German industries.


99+% of people’s time? I think you missed out web browsers. The majority of which are open source.

Not sure this also counts but almost 100% (leave a side asm.js/webasm for now) of the code run by browsers (JavaScript) also comes eith access to source.


The JavaScript that arrives at the browser is usually not source code in the way that licenses tend to define the term (which is the preferred form for modifying the program). It's usually at least minified.

It's kind of an interesting situation since it looks like source code even though it really isn't.


Looking at the source code of nonfree Javascript is not recommended. You could be sued if you happen to write something similar and can't claim anymore that you invented it independently ("white-room").


That's not really true - most JavaScript is minified and a lot is obfuscated.

And a lot is transpiled from another language. I'd say very few major sites send you readable hand written JavaScript.


Open source does well when companies build commercial products on top of it and then pay the developers.


I agree with you there. I love open-source, I barely use anything else. And yet I have to admit that much of it is absolutely not user-friendly in the sense of "I can explain this to my grandmother in under 10 minutes".

Much of it is technically amazing, and absolutely fantastic for technical users, but not really suited for the general consumer market. (With some notable exceptions, of course - see Firefox.)


The debate around copyright today is just so weird I think;

"I should be able to prevent anyone from producing anything too similar throughout my entire life, and my heirs should be able to stop them for another lifetime after I'm gone."

"Why?"

"Because the money they make is money I could've made."

"So, make another work that you can monetize?"

"But why would I work if I have to work?"

I can understand copyright to a degree, but the lengths we've agreed to in the West is absolutely ridiculous and ripe for abuse by companies that are built on aggregating IP.


Agreed, and I also think this investment/lottery/entitlement mindset is toxic because it devalues the actual work.

Human progress, our progress, comes from the work being done, not the money that can be made from it.

While money also end up funding other work, it's blindingly clear through the distribution of money and the proliferation of various speculation markets, that actually doing work to earn money is almost seen as 'earning' it in a roundabout way.

All you really need to do if you have real money, is to do nothing at all. Because while you can earn even more by doing something, even absolutely safe positions will get you a higher lifetime income increase than the average worker can expect over their lifetime.

It's simple math that shows how extremely we've devalued work, and I believe it has a negative effect on lots of things, some that might even too obvious to be visible at all.


see also the financialization of the economy, where money chasing money is far far more valuble than building actual real things, in the real world.


> and ripe for abuse by companies that are built on aggregating IP

Guess who pushed the copyright laws to where they are now ...


>"So, make another work that you can monetize?"

How are they going to monetize it if the printing presses will just print the work for free?


That’s what I mean by ”to a degree”. If you get a few years of time with copyright and time to monetize it tgat’s not too stifling for creativity. If you however get two lifetimes that’s just silly.


Cambridge economist Ha-Joon Chang’s book “Kicking Away the Ladder” describes how loose IP regulation has been at the center of the economic development of many countries. Sadly, once countries have developed to a certain level they tend to implement stronger IP to protect their position against other countries.

http://www.paecon.net/PAEtexts/Chang1.htm


Isn't it the same case for the United States in the 19th century? There were copyright laws protecting the states, but these didn't extend internationally, so books and other material could be brought stateside and be disseminated or published without punishment.


International works were not protected by copyright in the US, but it didn't really matter because of how long it took for copiers to get their copies on the market.

The vast majority of books made most of their sales shortly after publication. By the time the copier could get a copy of the published book, typeset their edition, and go into production almost everyone who wanted the book had bought a copy from the original publisher.

Stephen Breyer, back when he was a law professor at Harvard rather than a Supreme Court Justice, wrote an article in the Harvard Law Review called "The Uneasy Case for Copyright" exploring the idea that because of the limitations of copying technology much of earlier copyright law really didn't do much.

It's an interesting argument, but technological advances in copying have made it largely inapplicable for the present day.


Charles Dickens did disagree with this view: https://creativelawcenter.com/dickens-american-copyright/

But there are of course two sides to this story. Pirated editions back then as today increase the popularity of the artists, often more than their agency could do.

Unsurprisingly, even in the 19th century they came to a similar solution like our current music industry (minus the unsuccessful DRM part). You can't pirate live performances: "When he returned in 1867, he had mastered a way to generate alternative income streams based on his celebrity status. He performed readings of his work to packed houses and collected the receipts from the door."


> Pirated editions back then as today increase the popularity of the artists, often more than their agency could do.

Could work the other way too, though. Alexander von Humboldt had troubles with one of his books because a pirated translation into English was completed before the authorised version. Unfortunately, the pirated translation was so shoddy the book got really bad reviews in England...


It's also the reason for Hollywood - Film companies set up in LA to make it harder to enforce patents against them.


Hell, USA didn't sign the international copyright convention until 1980!


The historian claims that the missing copyright is "the real reason" for industrial expansion. I think it's "only" a contributing factor. Germany was divided in many smaller states and every state had their own universities. Many of these states were competing with each other about talent. So instead of centralising everything they had a distributed scholary culture. That's probably why there was a culture of "thinkers and readers". EDIT: Hence why a copyright law was not really enforcable. So the root cause was the division and competition, the missing law just a symptom of it IMHO.

Plus: lot's of iron and coal easily accessible and many big commercially usable rivers for transportation.


Suppose we even abolished all patents. Would it even make a difference in the software world? The selfish already got their new tool - online services. Not only you never have to release any source code, the user doesn't even run the compiled version himself. All he gets is output. The code running his computations may or may not be patented, he'll never know. I don't think licences like Aferro GPL can realistically be enforced.


I think patents are relevant when you do chemistry or biology research.

If you burn money for years designing a molecule, you could say you "own" the molecule, so you patent it.

The problem with the patent system is that it's necessary in a world where you need money to do any simple thing, because states cannot fund every research project, since it's impossible to know the outcome of every research.

But obviously you cannot patent everything, and that's where the patent system is failing. There is no good legal definition of a worthwhile invention.

Software is not patent-able because it never really is an invention that requires to burn a lot of money, unlike other domains of research. Anybody can write any kind of software.

But patents are still relevant where research requires expensive resources.


The original idea with patents was that they should make it worthwhile to invest in innovation and otherwise expire as fast as possible so that eventually all profit. Our current patent system has forgotten that second part and morphed to "expire as late as possible", which only serves greed.


The fashion industry is another example of an area that has thrived because of a lack of copyrights, as argued is this TED talk :

"Copyright law's grip on film, music and software barely touches the fashion industry ... and fashion benefits in both innovation and sales, says Johanna Blakley. In her talk, she talks about what all creative industries can learn from fashion's free culture.."

https://www.ted.com/talks/johanna_blakley_lessons_from_fashi...


To me copyright has a lot of similarity to GDPR. I posted about that now actually: https://news.ycombinator.com/item?id=17330914


In an abstract sense, you may be right.

In practice, authors sign away the ownership to a publisher in a profit sharing deal and it ends up protecting the publishers profit.

Makes me wonder of there is a business model for protecting user data, by managing any user-business data relation/permission and hunting down any company who is inappropriately using data from any member and collecting the fines. That would be the first protection racket that I'd like...


> The German proliferation of knowledge created a curious situation that hardly anyone is likely to have noticed at the time. Sigismund Hermbstädt, for example, a chemistry and pharmacy professor in Berlin, who has long since disappeared into the oblivion of history, earned more royalties for his "Principles of Leather Tanning" published in 1806 than British author Mary Shelley did for her horror novel "Frankenstein," which is still famous today.

In the absence of copyrights, and with plagiarism being rampant (as the article claims) how could authors command favorable royalty agreements? Could a rival publisher not just print the same book, sans royalty agreement, and undercut the original publisher?


The printers would have been able to sell fresh material, and perhaps the payments to the authors were insignificant if the volume was high enough. Competitors would also have had to replicate the work of typesetting and printing, which would take a while.


I remember reading that the strength of manufacturing in China is also that everything gets copied and innovation spreads quickly.


Process innovation and best practices is not generally something that gets IP protection.

AS for the article: The richest countries have strong IP protections, generally, that might indicate something.

Certainly creative trademarks, brands etc. - I don't like Disney but I think they have the right to protect Mickey Mouse so long as they are a viable entity/business.

Drugs ... this is one place where innovation might come to a halt of things could not be protected. I'm aware Unis do a lot of key research, but bringing something to market is hugely expensive.

Also remember a lot of patents are defensive in nature - their only purposes is to prevent trolls. So that's a sad artifact of it all.

No doubt the system needs an overhaul.


"The richest countries have strong IP protections, generally, that might indicate something."

The assumption here seems to be that their riches come from IP protections. It could be argued that the IP protections are a result of lobbying from rich companies.

I am not arguing that the causation is wrong, but the implied claim does not follow from the observation.


Or it could be the causation is correct, but for a rather unsavory reason, the wealth came from economic rents due to IP protections. The argument that IP defenders need to make is that these innovations could not have happened without strong IP protections, and I’ve never seen a compelling argument for this.


> I don't like Disney but I think they have the right to protect Mickey Mouse so long as they are a viable entity/business.

I strongly disagree with that. Just like with patents, the regulative protection must end at some point so the cultural and technological heritage is shared among all descendants, and not just a few lucky. The protection duration is already too long, not too short.


>AS for the article: The richest countries have strong IP protections, generally, that might indicate something.

you have something in the line that correlation is not causation?


Patents laws were in place before the industrial revolution, in the UK long before, and in the US around mid 19th century.

So most industrial growth happened after patents were part of commercial practice.

I suggest patents are less important now than they were then at least for what they are supposed to be used for ... patents may be increasing in terms of corporate importance, but more as a general legal asset than anything. Trademarks and copyrights is a whole other thing.


patents laws and their actual enforcability is something else entirely


Case in point - US aviation being backwater until Rights patent expired while the rest of the world saw explosive innovation in aviation.

Patents were necessary only when public disclosure had value. In the world where the same things get developed almost simultaneously by different people at different places and anything can be quickly reverse engineered the public gets no value in return for granting monopoly through a patent.


What I really liked about it is how it forces publishing companies to truly make knowledge available for everyone. It's the direction music is going in this country, where individual listens are essentially a free public resource for anyone who pays a minuscule subscription fee.

Books and movies seem to be retaining pricing power for some reason. I feel like the ultimate reason behind this is that books and movies take a lot more time and attention to consume.

I know for me, I would be far more inclined to read and watch more of them if it were easier to break them up into chunks. Give me a specialized movie player that lets me pause and start movies by scene, make individual books and movies super cheap, sure, I don't mind paying cable-like fees for something actually useful and I'd watch movies a scene at a time by the dozen. Give me a feed reader that mixes in book chapters, in order, with my normal news feed.

Essentially, I don't want to waste time on crappy platforms. Just give me the damn content in a way that makes sense for me to use it. I may get to the point where my appetite for information is so voracious, I'll just mass pirate stuff and upload it all into a personal database so I can chunk it myself.


> What I really liked about it is how it forces publishing companies to truly make knowledge available for everyone.

I haven't read the article yet, but wow will that interact with: https://www.eff.org/deeplinks/2018/06/eus-copyright-proposal... [0]

[0] https://act1.openmedia.org/savethelink


Interesting point about publishers producing a cheap and a "classy" version of their books already way back then. I seem to remember reading somewhere else that when paperbacks first came up, they were a huge success, yet were criticized by the more snobbish educated elite as a virtual blasphemy...


Tolkien at least had that attitude towards LOTR being printed in paperback in USA in the 60s (never mind that the publisher did so on questionable legal grounds).


This was to compete with the pirated editions - its mentioned in the forward of my UK edition.


I have a paperback version of LOTR in a single volume. Very convenient. Has the full text in relatively small font in just under 1200 pages.


If there’s one argument against copyright I would bring up it’s patent trolls. If you can propose an IP law to me that makes sure creative individuals are not bullied and stifled by unscrupulous people hungry for a quick buck I would be more inclined to support copyright laws.


And the unification, industrial revolutions, Bismark and all those wars had nothing to do with it - seems a very narrow view of 19th century German history


I see the article more as pointing out one aspect that is often ignored - not as invalidating all those other aspects.


Wonder what kinds of software we have now if we didn't have Copyright Law for source code too?


There is no justification for copyrights and patents any more than there is for any other monopoly.

However if we get rid of them there is a need to socialise creative production so that artists of all shapes and hues in all creative fields can still afford to live simply doing what they love to do.

It's a fair trade. Producers can copy and extend whatever any human has created but in return they have to produce enough to look after them.


>Producers can copy and extend whatever any human has created but in return they have to produce enough to look after them.

Which is just a royalty with extra steps.

The problem with most attempts to subvert a capitalist/market economy is that they all only manage to obfuscate the usage of money and ownership and replace a lot of it with centralized control.

Your idea still requires ownership of IP and compensation for it.


Rentier activity, as usual, inhibits growth.


What means no copyright law? Of course we have laws about it.


> Did Germany experience rapid industrial expansion in the 19th century due to an absence of copyright law?

That's the very first sentence...


Ah and 2010 is in the 19th century? I guess I was never good in math.


That is the standard way to write that the article was published in 2010.


> Authors are only motivated to write, runs the conventional belief, if they know their rights will be protected.

This is absurd. How stupid does one have to be to think this? Copyright serves only one purpose: to make profits for the author and, more importantly, to make profits for the publisher and distributor by exploiting the author's work. These days it almost exclusively does the latter. Copyright is not about contributing to or advancing society. It has nothing to do with that regardless of the original intentions behind it. Copyright is about making money for a few individuals or groups at the expense of the wider society. The sooner the idiotic conventional thinking quoted above is replaced by a proper analysis, the closer we are to fixing this atrocious system that's holding us back.


"No man but a blockhead ever wrote, except for money." --Samuel Johnson

http://www.samueljohnson.com/writing.html




Registration is open for Startup School 2019. Classes start July 22nd.

Guidelines | FAQ | Support | API | Security | Lists | Bookmarklet | Legal | Apply to YC | Contact

Search: