All posts by Jakke Mäkelä

Physicist, but not ideologically -- it's the methods that matter. Background: PhD in physics, four years in basic research, over a decade in industrial R&D. Interests: anything that can be twisted into numbers; hazards and warnings; invisible risks. Worries: Almost everything, but especially freedom of speech, Internet neutrality, humanitarian problems, IPR, environmental issues. Happiness: family, dry humor, and thinking about things.
Examiner of Silly Patents

Examiner of Silly Patents 3: Monkey thermometer

2 Comments

To take a break from all the gloom and doom of the last few week’s blogs, here is a different kind of look at patents. Very different.

Every now and then I will take a silly patent and pretend it isn’t silly. I will analyze and defend it in all faux-seriousness (see disclaimer at the end).  Usually I cannot defend it, but learn something in the process anyways.

From 1987, US 4,634,021 (Release mechanism) has a self-explanatory picture.

“A release mechanism is disclosed for releasing an object such as a ball from a body under the force of gravity. A bimetallic element obstructs or opens an opening in the body for retaining or releasing the object depending upon the temperature of the bimetallic element. The release mechanism may be incorporated into a novelty “brass monkey” for “emasculating” the monkey when the temperature decreases to a predetermined temperature at which the balls in the “brass monkey” are permitted to drop to a base which is designed to produce an audible sound when struck by the balls”.

I always try to give positive feedback and constructive criticism. In this particular case, the positive feedback is immediate and obvious. Anyone with the self-confidence to eternally attach his name to a drawing like this deserves our respect. Strength through goofiness.

On the constructive criticism side, there is more to be said. I believe that the inventor did not choose an optimal strategy to protect his excellent idea of using an emasculated monkey as a thermometer.

In my view, by patenting the inventor divulged too much information and gave the competition an unnecessary advantage. He opened his strategy, without really protecting it. Fundamentally, there seem to be too many workarounds around this patent.

For example, this patent most likely does not cover other animals. Castrating a donkey to tell the temperature would almost certainly be possible despite this patent. Using a human being would potentially lead to a court case. The arguments would revolve around whether a human being is biologically simian enough to be considered a special case of a monkey. I hesitate to speculate how that court case would end.

There are also non-testicular extensions of this idea which could have been pursued in the patent. In particular, something like a system with a French Revolution theme could present a similar user experience: when it gets cold, Marie Antoinette’s head is chopped off with a clang. This patent does not prevent such a user interface from being implemented.

There are also technical workarounds.  A bimetallic temperature valve is well-known. However, if the bimetallic valve were to be replaced by a sphincter-like structure, it is probable that the patent would not cover it.

What about the business case?  It is possible that this patent has indeed hit a niche which has not been extensively filled. I did not really find anything in the patent literature that would imply that this is a major technology area.

A Google search for “novelty thermometer” shows examples of today’s state of the art. I believe the figure below has some commonality in spirit with the patent. A rectal thermometer in a duck is used to measure bath water temperature. It is an unexpected combination.

Source: Screen capture.

I believe that a design patent to cover just the monkey implementation could have been more cost-effective.  Design patents are cheap to file and have no maintenance fees; on the other hand, they are only valid for 14 years against a utility patent’s 20 years. Alternatively, the inventor could have tried to broaden his patent significantly to cover other animals and user interfaces. That would, however, have raised the costs.

I strongly suspect that the inventor did not manage to make money from this patent. However, that does not in any way detract from the inventive step of this patent; one would not expect the see a thermometer implemented by using the clangs caused when a brass monkey is neutered and its testicles fall off.

Although an engineer type should never venture into aesthetics, I want to say something about artistic values. I find something poignant yet majestic about the figure of the monkey. This is almost worthy of a monument. If I had been the inventor — though I may be alone in feeling this — but if I had been the inventor, I think this figure would make an excellent gravestone. Why not? What could be a better memento for future generation to remember me by? Strength through goofiness is something to celebrate, throughout the generations.

All Examinations of Silly Patents: click here. 

Disclaimer: these analyses have very little to do with anything, and in particular have nothing to do with legal issues. Most of the patents cited are expired (or should be). I do not touch the “claims” section, which is the legally relevant part. These blogs constitute prior art, so that any new any ideas expressed here can no longer be patented.

Patents, SMOS

SMOS: Humanitarian Patent Pool

3 Comments

What would it take take to actually, truly start  a “Humanitarian Patent Pool” like Timo Tokkonen suggested in a posting last week? The idea being to collect certain patents in a non-profit “pool” to keep patent trolls away from ruining humanitarian efforts.  The question was inspired by humanitarian catastrophe communications (see SMOS web page). But there could be other areas.

Is HPP even vaguely realistic? Patents are powerful, but not all-powerful: the Doha Declaration allows developing countries to bypass existing patents for medicines when public health is threatened. A concept similar to HPP, defensive patent aggregation, exists in the commercial world, but to my knowledge not in a non-profit setting.

Here are some back-of-the-envelope estimates on how the HPP might work.  Bear with me if there are ludicrous errors, and please propose improvements.

The core purpose of the HPP must be to eradicate patent trolls in humanitarian areas. The purpose is not to hinder legitimate players. (This is immediately a controversial goal. Idealists will want to eliminate patents from the humanitarian field altogether. I feel it is sufficient to eliminate just the rabid dogs, and let the healthy ones thrive).

The HPP needs to be a non-profit foundation. It should be international in scope, but it might be sufficient to restrict it to the USA because that is the home of the patent trolls.

The key function of the HPP is to collect ownership of “bad” patents. By “bad” I mean something that is not being used to create anything, but can be used by a troll to stop development. A much more refined definition is obviously needed.

The main category of “bad” patent is one which is too broad and should never have been granted in the first place. Another category are patents for a technical solution which has become obsolete, but which a lawyer can stretch to cover some other technology. Such patents are lethal weapons in the hands of a troll. But they could also be lethal weapons in the hands of the HPP.

There are two key strategies.

  1. Containment and decay. Collect patents that are allowed to expire as soon as possible. The purpose here is to prevent trolls from making claims, and to create strong prior art against future spurious patents. These patents should for the most part be collected through donations.
  2. Active deterrence. Patents that can and will be used in litigation against trolls. In some cases, the HPP might consider paying for these. The cost of filing a patent is > 10 kEUR, so the HPP might be willing to consider buying them at cost.

The HPP should be aggressive, not defensive. Unless the HPP is willing and happy to go to court against trolls, it will have no deterrence effect.

Since the USA already has a well-oiled machine in the Electronic Frontier Foundation, it would make sense to tie the HPP in very tightly with the EFF, especially its Patent Busting Project.  On the other hands, in terms of funding, the HPP could have wider appeal and hence be independent.

The working principle needs to be absolute transparency. For strategic reasons, if preparing for an attack, the HPP may require secrecy. But even there things need become public when fight is over. As far as I can see, the HPP cannot ever accept secret agreements or settlements.

The focus areas for active deterrence must be chosen very tightly. They should be restricted to those areas in which humanitarian damage can be massive, and in which trolling activity seems to be particularly easy. Catastrophe communications would certainly be one such area.

The budget of the HPP simply cannot be kept small. Even if using only donated patents, there are legal costs associated, even if the patents are allowed to expire immediately. A reasonable minimum estimate is 1 kEUR per patent. Since there must be hundreds to thousands of patents in the pool, this easily results in a budget of hundreds of kEUR per year.

If some key patents are bought with hard money (though at cost), the cost per patent could be 10 kEUR. The number of truly crucial patents will be small — court cases are typically litigated over just a handful of patents — but knowing which ones a critical requires buying more.

The cost of stockpiling and maintaining the patents will easily climb to a million EUR per year. Unlike projects like the EFF’s patent-busting project, there is no meaningful way to crowdsource the idea. It needs hard money.

If the HPP goes to court, the legal costs are unpredictable, but the HPP must be able to handle them. This is where my reasoning becomes completely fuzzy. Could this work on a pro bono principle? If courts find trolls’ patents spurious, might they willing to force the trolls to pay costs? I find this highly problematic.

Is there any way for the HPP to make some profit to recoup all its losses? In principle, yes, by licensing to legitimate businesses. However, trying to make a distinction between non-legitimate and illegitimate players would add costs and make enforcement difficult.

And, more crushingly, the HPP would risk turning into a patent troll of its own. (“He who fights monsters should see to it that he himself does not become a monster”). It is difficult to see any way to make the HPP self-sufficient.

So where would the funding come from, then? I have no real idea. This was as far as I was able to get in one sitting.

So is the HPP idea even vaguely realistic? If not, would be worth developing from some other angle? If not, do we just have to adapt to life with the trolls?

 

SMOS

SMOS: The Kiss of Death of IPR

For once I will be blunt: I now feel that with crisis communications hardware, the first whiff of IPR (patent) issues will cause the effort to crash and burn.  My opinion on this was certainly different before we started the SMOS project (see SMOS web page).

The idea of combining humanitarian technology and intellectual property sounds uncomfortable. After all, humanitarian activities are supposed to be humanitarian. Nevertheless, there were perfectly valid reasons to consider IPR issues with SMOS.
1. Protect the idea in order to make at least a minimal living from it.
2. Prevent others from destroying or profiting extortionately from the idea.
3. Improving interoperability with telecommunications providers by playing with the same rules (rather than adopting a “hacker mentality”).

I am a deep skeptic about IPR, but do not consider it fundamentally “evil” — it may have its place.  However, I am coming  to the conclusion that for the specific case of crisis communications, IPR is the kiss of death.

The numbers

Since we pursued the possibility of patenting quite heavily, I learned some ballpark figures.

  • To file and get a Finnish patent: 6000-8000 EUR. Most of this goes to the patent agency handling the application. There is no practical way to survive without a patent agency.
  • In Finland, a decision is usually rendered within a year. There may be interim decisions which require using a patent attorney, pushing up the price.
  • After being granted, the annual cost of maintaining a patent is small for the first three years, but starts rising rapidly after that.
  • To file and get a European patent, the cost is approximately twice of a Finnish patent. Maintenance costs are also roughly double.
  • A US patent requires using a patent attorney in the US. Overall cost is comparable to a European patent. In the US, a decision can take 5-10 years.
  • I have no idea what the costs would be in countries such as China or India.
  • A figure of 20,000 USD (or EUR) per patent is often used a lifetime cost, at least within a single country. However, for a catastrophe communications system, protection would be needed in multiple countries, some without an advanced IPR culture.
  • → If I had to give a single number, it would be 50,000 EUR per patent.

Realize, however, that a single patent has no practical value. To protect an idea in any significant sense, a dozen or more patents may be needed. If the IPR route is chosen, a cost of half a million just to file patents is realistic.

Note also that filing patents does not in itself guarantee anything. Anyone can infringe on a patent, and it is up to the patent owner to sue. Even minor court cases can eat up six-digit sums of money, and can be decided on the basis of a comma in the wrong place, even if the most expensive lawyers are used. Going the litigation route means risking millions on what is in effect a roll of the dice.

Even worse, in this particular case litigation is a no-win situation. If the other side is a local company in a developing country, it can (and most likely will, and perhaps should) take the role of a victim being bullied by large Western IPR interests and inhibited from trying to save lives. There is no real way to protect one’s public integrity in such a case.

We had one major rationale for the IPR route: we felt that by making hardware along commercial lines, it would be easier to arrange interoperability with operators. I am now highly skeptical about this. Interoperability would simply cause operators to view SMOS as a potential competition. It would take a major player 15 minutes to dig up enough patents from its patent thicket to make our life impossible. Whether those patents are relevant is completely irrelevant. The threat is enough.

The implications

I draw a harsh conclusion: any whiff of IPR will cause a humanitarian crisis communications project to crash and burn.

On the other hand, proceeding without a thought for IPR is troublesome for funding, since venture capitalists want to recoup their investments within a few years. Without IPR, there is less to recoup.

In the traditional way of thinking, having no IPR protection would be unthinkable because of the competitive issues. However, I believe I have demonstrated above that IPR does not actually give any benefit in this specific technology case.

What to do about this then? I am certainly not saying that crisis communications hardware projects are hopeless. However, it would be wise to pursue such projects with the (perhaps unwritten) attitude that IPR is the enemy.

An open-source approach seems appealing, and has been done successfully with software (for example Sahana). But hardware?  It is only possible to go so far with a “hacker mentality”; if there is a desire to use cellular telecom interfaces, then there is a need to cooperate with commercial providers as a commercial entity.

One of the few ways to actively defend a project of this type against patent trolls is to have a strategy of defensive publishing. In other words, publish all ideas as soon as someone blurts them out, ideally in the form of a permanent Creative Commons-licensed blog. This means that the blog becomes prior art, and the ideas can no longer be patented by anyone.

Other than that? Perhaps the best advice I can give is to look at the material on the SMOS project page, study what we tried to do in SMOS, and then try to not do that.

SMOS

The SMOS project

When disaster strikes, could we airdrop small base stations and set up cell phone communications that way?  

See the previous posting to see where that simple question took us. (The post is long but entertaining).   Kalle Pietilä asked the question  in early 2011. Answering that question became an off-time project that occupied four of us (Timo Tokkonen, Jakke Mäkelä, Niko Porjo, and Kalle Pietilä) for many evenings and weekends in 2011-2012. The previous posting described the process; this posting summarizes the idea. See the SMOS project web page for technical information.

We named the project SMOS, for “SMS Our Souls”. The core of the concept is this: by cutting away all functionality except for SMS traffic, could one develop base stations so small, robust, and low-power that they could literally be sown from the sky in case of a major catastrophe? This extremely tight specification would mean that even if the modules are not actually airdropped, they will be small enough to be hand-carried anywhere by volunteers.

We pursued this as a potentially realistic commercial project, in our off-time and under the radar. We developed sales material, technology concepts, use cases, and business calculations, pitched the idea to various places, and looked for funding sources.  By early 2012, it became clear that no traditional start-up business case existed. We decided to abandon the project (at least for the foreseeable future) due to reasons described in the previous posting.

We were then faced with two choices: pretend we never tried and hide our “failure”, or publish what we did find in case someone else finds it useful. We opted for the latter.

We are releasing the relevant material under a Creative Commons license (CC-BY), which means that others may use the findings freely (including commercial use).

Project web page: Click here.

Since we had discussions with some commercial partners, some information needs to be redacted because of confidentiality and trust reasons. Also, we are still a little behind on  cleaning up some of the technical material,  but we will get there in a week or two.

Just as we were putting the finishing touches here, we heard of a new startup called Tethr, which is trying to do very similar things. Please read the BBC article on them, and do join their mailing list. We have great hopes for them.

Below is the text of the one-pager  (pdf: SMOS 1 One-pager) that we used to open discussions on the idea. No real solutions are proposed here, just questions.

THE “SMOS” SOLUTION: ONE-PAGER

After any natural or man-made catastrophe, the first few hours and days are crucial. Before real recovery efforts can begin, it is necessary to set up a communications system to coordinate the efforts. Disaster communication systems exist for emergency officials and aid workers.

For normal citizens, mobile phone networks are becoming an increasingly useful way of communicating, as well as being a potential information source for rescue efforts. Just knowing the number of active mobile phones can allow the authorities to focus rescue efforts. Even developing countries have high mobile phone penetration, and thus the number of phones is a reasonable proxy for the number of victims. Although the batteries of mobile phones will be depleted, for the first few days the majority will still be active.

However, even mobile phone networks can be destroyed or overloaded when the catastrophe is severe enough. The SMOS Initiative is proposing a solution to enable partial mobile phone connectivity even when the normal infrastructure is not working. More specifically, it is simpler to restore text messaging (SMS) services than full voice service. An SMS service would allow victims to communicate with their families. At least as importantly, it would allow emergency organizations to receive and send information straight to the victims.

If it is accepted that the temporary network only transmits SMS traffic, then the required hardware could be simplified and miniaturized compared to normal cellular telephone base stations. The target of the SMOS Initiative is to define a “base station” small and rugged enough to be aerially deployed, so that one volunteer with a private pilot’s license and a small airplane could quickly restore telecommunications over a huge area. Even if aerial dropping is not possible, the small size means that they are easily portable by other means.

Once the base stations are dropped, they automatically form a network for mobile phones to transmit and receive SMS traffic. The base stations are powered by batteries that give the station a lifetime of about a week (after which normal wireless infrastructure is assumed to be in place).

The proposal still has multiple open technical issues, and the economic, logistical, and political challenges are also formidable. The idea is therefore presented as a discussion paper, aiming to open debate on whether such a system could in fact be practical and useful.

Note (April 2012): The project has been discontinued in its present form, as no commercially feasible way of implementing it has been found.

Päivän heitto

Re: Miksi liikennejärjestelmässä luotetaan sääntöihin?

4 Comments

 Voisiko syntyä jotain kaunista, jos yhdistettäisiin teknologia ja aasialainen elämänasenne? Nöyryyttäminen on turhaa, jos teknologia on inhimillistä.

Niko Porjon eilisen kirjoituksen innostamana havahduin miettimään autoiluasiaa tarkemmin. Kirjoitus on provosoiva, mutta lukemisen arvoinen.  Porjo kysyy: miksi vahinkojen pitäisi autoilussa olla rangaistavia? Vahingot ovat vahinkoja, niitä tapahtuu aina kun ihminen jotain tekee. Miksi niitä silti autoilussa kohdellaan kuin rikoksia? Porjo tiivistää asian omalla tyylillään: “Pidän koko ajatusta liikenteen toimimisesta rangaistuksen uhan voimalla kuvottavana.”

En taida lopultakaan faktoista olla Porjon kanssa samaa mieltä, mutta kirjoitus sai ajattelemaan. Siinä on jotain ideaa.  Samalla tajusin, että juuri autoilusta on vaikea keskustella järkevästi.  Autoilu on asia, johon Suomalainen Mies suhtautuu tunteella. Se on miehuuden ytimessä. Vapauden. Suomalaisen elämänmuodon. Omakotitaolasumisen. Kesämökkien. Eikä toisaalta pidä unohtaa, että autoilussa on aina pelko mukana.

Siksi yritän löytää sovellusalueen, jossa olisi helpompi pysyä rationaalisena.

Ymmärrän pohjimmaisen argumentin näin: Porjo ei väitä, että rangaistukset pitäisi poistaa.
Sen sijaan liikennejärjestelmiä tulisi kehittää niin, että tahattomat virheet olisivat (lähes) mahdottomia. Porjo esittää yhden teknisen idean — puomit valoristeyksissä — joka on selkeästi keskeneräinen, mutta antaa konkreettista tuntumaa ajattelutapaan. Keskikaiteet ohituskaistojen kohdalla ovat selkeämpi esimerkki; vaarallisia ohituksia ei pääse tapahtumaan, koska ne on fyysisesti estetty.

On makukysymys pitääkö tätä realistisena vai ei; itse en pidä, koska se vaatisi liian massiivisia muutoksia koko kaupunki-infrastruktuuriin. En kuitenkaan avaa  autoilukysymystä sen enempää, vaan sovellan ajatusta toiselle alueelle: plagiointiin (J. Järveläiselle kiitos ideasta).  Plagiointi on todellinen ongelma yliopistoissa, ja kouluissa, ja tätä vauhtia kai pian päiväkodeissakin.

Perinteisesti ajatellaan rankaisua: käytetään erilaisia tunnistusohjelmia (täällä, täällä, täällä, ja ympäri Internettiä) löytämään plagiaatit, ja isketään sanktioita kun joku jää kiinni. Euroopassa jää säännöllisin väliajoin joku poliitikko kiinni väitöskirjansa plagioinnista; heidät erotetaan ja heitä nöyryyttää koko maa (sivumennen sanoen, aina kun kuulen tuollaista, käyn halaamassa omaa väitöskirjaani ja toivon ettei kukaan koskaan huomaa kaivella sitä tarkemmin. Mutta jos joku haluaa kokeilla, väitöskirja löytyy täältä).

Mutta miksi? Aidosti tahaton plagiointi on uskomattoman helppoa nykyisillä tekstinkäsittelyohjelmilla. Leikkaan tekstiä toisesta artikkelista, tarkoituksena muokata se myöhemmin; joku soittaa; muokkaaminen unohtuu. Sitä vain kerta kaikkiaan tapahtuu. Tai lukiolainen kirjoittaa esseetä, eikä yläkoulun nukuttuaan ole vieläkään sisäistänyt millä tavalla Wikipediaa voi lainata.

Lukiolaisen tapauksessa  nelonen kyllä sisäistää asian tehokkaasti ja kerralla.  Mutta miksi sisäistämisen pitäisi tapahtua nimenomaan nöyryytyksen kautta? Miksei plagioinnintunnistus voisi vaikkapa olla osana tekstinkäsittelyohjelmaa, samalla tavalla kuin nyt on automaattisia spellcheckereitä? Tällainen automaattinen tarkastus poistaisi (periaatteessa) tahattoman plagioinnin mahdollisuuden.

Tämä on kaikkea muuta kuin pitkälle pureskeltu idea, ja mm käyttöliittymä ei ole todellakaan helppo. Miten ohjelman pitäisi plagioinnista varoittaa? Miten ilmoituksen voi ohittaa? Miten kirjoittaja voi todistaa, että on käyttänyt ohjelmaa oikein? Miten toimitaan tapauksissa, joissa “plagiointi” voi olla jopa sallittua, esimerkiksi väitöskirjojen tiivistelmissä? Paras ratkaisu on ehkä triviaalimpi: vaaditaan opiskelijoita itse ajamaan kirjoituksensa plagicheckerin läpi ennen jättämistä. Mutta olennaista on se, että ongelma on hyvin rajattu, ja  teknologia on jo olemassa.

Jos ohjelma on tehty oikein, plagiointi on mahdollista vain jos kirjoittaja tekee sen tietoisesti. Tällöin, ja vain tällöin, sanktiot ovatkin paikallaan.  Uskoakseni filosofia on samantyyppinen kuin Porjon. Miksi käyttää nöyryytystä, jos pehmeämmätkin keinot ovat teknisesti mahdollisia? Vähennetään tahattomien virheiden mahdollisuus lähes nollaan, jolloin lainvartijoiden panokset voidaan laittaa oikeaan paikkaan: törkeiden tekojen tutkintaan.

En ole vielä miettinyt läpi, kuinka laajasti tätä ajattelutapaa voisi soveltaa. Se sotii hiukan suomalaista brutaalia perusluonnetta vastaan. Mutta tässä voisi hyvin ottaa oppia aasialaisesta kulttuurista. Jos asiat voi teknologialla hoitaa niin että kaikkien kasvot säilyvät, niin miksi ihmeessä ei hoidettaisi?