Flickus flackus flum

Jacobs tankar om och med kvalitet

Archive for January 2016

Utställningen om Gårdaalternativet

written by jacob, on Jan 29, 2016 11:45:00 AM.

I går var jag och tittade på utställningen om Station Gårda på stadsbiblioteket i Göteborg. Det är ett förslag på hur kollektivtrafiken i Göteborg ska lösas, som ett alternativ till Västlänken. Det senare är ett projekt som till varje pris ska drivas igenom av en koalition mellan borgerliga och rödgröna partier i kommunfullmäktige, trots ett kompakt motstånd från stadens medborgare.

Efterhand som planerna på Västlänken har börjat bli mer och mer konkreta så har problem och frågetecken hopat sig. En enorm prislapp på 20 miljarder kronor i 2009 års penningvärde. Mycket liten nytta, eftersom resandeunderlaget till stationerna i Haga och vid Korsvägen inte är speciellt stort. Ingrepp i känsliga miljöer, med stora risker för sättningar i kulturellt värdefulla fastigheter. Kaos och avspärrningar av stora delar av Göteborgs centrala delar i tio års tid. Negativ samhällsnytta. Tekniskt besvärligt projekt, med långa sträckor i lera, skapar stora risker för spräckt budget och spräckta tidsramar. Komplicerade och otestade lösningar för att hantera tekniska problem kring Götatunneln och Hagastationen. Listan kan göras hur lång som helst.

Oavsett vad som har framkommit har de styrande politikerna i Göteborg vägrat att göra en total omprövning av projektet

Jag blev därför mycket imponerad av vad Gårdagruppen har åstadkommit i form av ett förslag om att bygga en helt ny centralstation för Göteborg i Gårda och att ansluta den med ett kompletterad spårvägsnät och en helt ny snabbspårvagn i en ringlinje runt Göteborgs centrala delar. Ringlinjen är tänkt att gå från Gårda till Gullbergsvass, därefter korsa Gäta Älv till Tingsbergsvass, gå på Hisingssidan till Lindholmen, korsa älven igen till Stigbergstorget, gå i tunnel till Linnéplatsen, Sahlgrenska och Chalmers, för att därefter gå tillbaka till Gårda via Carlanderska.

Man har noga utarbetat hur arbetet ska gå till i olika etapper, på ett sätt som minimerar störningarna. Man börjar med en ny tunnel förbi platsen för den nya stationen. Den ska återanvändas för helt separerad godstågstrafik efter byggperioden. E6:an får en tillfällig sträckning där stationshuset ska ligga medan man bygger en överdäckning av nuvarande E6. Därefter kan man börja bygga stationshus och tunnlar för perronger i berget. Stationen blir användbar redan när man har det första perrongparet på plats. Utvidgning med ytterligare perronger kan göras medan stationen är i bruk. Först när stationen är komplett stänger man ned nuvarande centralstation och kan då återanvända marken för att utvidga Göteborgs centrum.

Det här är bara några av alla detaljer som redovisas i utställningen. Det jag slås av är hur metodiskt man har gått över alla detaljer i sin planering och hur man har tillämpat ett systemtänkande, där stationen integreras med stadens kollektivtrafik, för att serva både gamla stadsdelar och de områden där Göteborg kommer att växa under det närmsta halvseklet.

Nu kanske det låter dyrt och extravagant att bygga snabbspårväg och en masa extra spår för att komplettera det reguljära spårvägsnätet, men genom att det är så mycket billigare att bygga i berg än i lera så blir kostnaden för station Gårda plus alla spårvägsinvesteringar ungefär lika dyrt som Västlänken.

Om jag fick välja så skulle jag alla gånger föredra Gårdaalternativet. Det är ett visionärt förslag som löser många av Göteborgs transportproblem, till skillnad från Västlänken, som låter ett fåtal människor från kranskommunerna enkelt ta sig till Handelshögskolan och Svenska Mässan.

Om du kan, gå och se utställningen. Skriv på listan som kräver en omprövning Västlänken mot Gårdaförslaget. Prata med representanterna för Gårdagruppen. Du kommer att inse att det är en samling erfarna proffs som faktiskt begriper mer om stadsplanering och komplicerade byggen är Västlänkens projektgrupp.

Shakedown in the Bitcoin world

written by jacob, on Jan 18, 2016 5:47:39 AM.

A few days ago Mike Hearn made a blog post, where he left the Bitcoin project, quoting a number of of problems with the Bitcoin community and the implementation of Bitcoin as a currency. Mike Hearn has been a quite active core developer for a number of years, so his departure is a notable event. A number of other people in the Bitcoin community have claimed that Hearns departure and his blog post were due to him joining a competing project run by traditional commercial banks.

While it may very well be true that Hearn is joining another project, his criticsm needs to be taken quite seriously by anyone taking an interest in Bitcoin. It is interesting to see that the major point in Hearns blog post echo the major weaknesses I pointed out in my review if Bitcoin (in Swedish) some months ago.

The blockchain

It turns out that changes to improve the security and reliability of the blockchain has lowered the practical transaction limit to 3 transactions per second, instead of the 7 I wrote about in my review. This is really bad news, as the initiative to raise the block size has stalled/failed. The current situation is that there are more transactions being performed than can fit in the block chain. This results in transactions with no fees or low fees being left out by the miners, resulting in long delays before they get executed. With even worse congestion, some transactions may never get executed. Bitcoin enthusiasts claim that this is not a problem. Small transactions will go away, since the fee to get included will be prohibitively high. There are two things that are wrong with that way of reasoning. The first one is that you are no longer a currency if small payments can’t be carried out. There are many merchants that have signed up for Bitcoin expecting to be able to receice small payments. They will turn elsewhere. The second one is that with congestion, the time that it takes to get a payment executed becomes longer and in practice indeterminate. Since the time it takes to make sure a transaction has gone through is 10-30 minutes to begin with, the delays can become major practical problems for both payers and receivers.

There is a solution in the works for the congestion, that is hoped to be in production by the end of 2016. This solution aims at separating out the signatures in the payments from the blockchain, reducing the size of each transaction to about a quarter. This is of course welcome, but I see two major problems. The first one is that a year is a very long time on the internet. While the blockchain hits its capacity ceiling, there will be a lot of uncertainty around Bitcoin and adoption of cryptocurrencies may go to more promising projects. The second one is that an increase of the capacity to 12 transactions per second is pitiful. Bitcoin is hitting the capacity ceiling when it is in an early experimental phase. I’d say that an increase in capacity of two orders of magnitude immediately is what is needed, and in the longer run, several more orders of magnitude must be possible. If the Bitcoin community can’t deliver that, I think the currency has a bleak future.

The miners

Hearn sees a problem in the concentration of miners in China and in the conservative stance of the miners. I see a much more fundamental problem in the mining process. As things has developed, a small number of miners with specialized equipment are providing the essential service of gathering transactions into a block and calculating the cryptographic signatures that are needed to secure the block. They do so in competition with each other, having to solve a puzzle whos difficulty is proportional to the amount of resources the collective of miners are throwing at the problem. This requires constant upgrade of equipment and high bills for electricity and cooling. Essentially, what the miners are doing is turning fiat money into heat, whith bitcoin as a byproduct. As they produce Bitcoin, they are increasing the money supply, making all other Bitcoin holders poorer.

Calculating the next block on the chain is an essential service, but it could be provided in a much less wasteful way. This would entail some sort of community agreement about what parties to entrust with the work and how to pay them, instead of the current anarchocapitalistic race for the maximum number of hash-calculations per time unit. The current proof-of-work scheme is not in any way more secure than a distributed network of custodians, selected by the community to perform the job.

Governance

It is quite obvious that the governance of Bitcoin is lacking. There are a number of interest groups that are not talking to each other and there is no process for reaching decisions. This is of course what is to be expected in a community with anarchistic ideals, but nevertheless a huge problem. Had Bitcoin been perfect from the start, it might have worked, but in the face of severe shortcomings, a faster and fairer way of making decisions is needed. As venture capitalist Vijay Gupta observed, Bitcoin contains all the elements of a voting system, where ownership of one bitcoin gives you one vote. Indeed, blockchain technology would be very suitable for the registration of stock ownership in companies, with voting rights tied directly into the application.

At any rate, voting rights does not a community make. You still need community representatives that have the trust of the community. The current situation, with a few developers with the power to push things forward, a few miners with the power to hold things back and a large proletariat of Bitcoin owners with no power at all is severely unbalanced and one of the sources of problems for the community. I think a governance body with representatives from all stake holder communities to take care of the daily business, and to put major issues to the vote in the community would produce much better results.

Satoshi Nakamoto

The anonymous creator of Bitcoin is actually a major problem for Bitcoin as a currency. He mined about a million Bitcoin in the early days, and they are still unspent. Since the cap for the total number of Bitcoin is 21 million, he will end up having just under 5% of the total number of the currency ever existing. Since a fair amount of Bitcoin already are lost forever, he may end up with more than 5% of the actually tradeable total. Having one individual controlling such a large part of the total supply makes me very nervous. Let’s say that Bitcoin is really successful and reaches $100 000/btc, and Satoshi decides that it is time to head for the door. I think it would result in a major panic and people losing fortunes (people would know that he was selling, since all Bitcoin are traceable from their creation to where they currently are). You’d be a lot safer if you knew who Satosho Nakamoto is, and what his motives are, but as things stand, you don’t even have that.

Conclusion

I think Bitcoin has a period of decline ahead. The current problems will take time to fix, if they ever will get fixed. If the community manages to organize and to fix the problems, I think the period of decline may pass. The market capitalization of Bitcoin is so much greater than any of the minnows in the crypotcurreny pool. However, it requires speed and flexibility, otherwise a currency that scales better and that has a better model of governance may take over the leading position. If that happens, Bitcoin will go the way of Betamax.