Been thinking about Thingamy again, and a few choice elements have come together that have actually provoked me to write about it.
Speaking to Hugh last night at the airport about what all is going on, and he mentioned Sig and Thingamy, which seems to be going well in the background. He said again, Sig doesn't really think that SAP is the main thing to replace, and I agree, but perhaps I should explain why.
The other thing that kicked off this line of thought was seeing this comment on Jerry Pournelle's web page.
"Subject: singularity conference
I think the singularity conference missed the elephant. While the number of transistors in a computer chip continues to double every eighteen months, the work produced by those chips stopped doubling some time ago. The AI humans have to be concerned about isn't in a box, it's in our corporate business structure.
At the dawn of the twentieth century the US Supreme Court ruled that corporations had all the rights of a human being, but few of the responsibilities. Marry that to an amoral legal structure that insists upon a separation of church and state that it enforces by recognizing only secular humanism as an acceptable world view and we have a system that has removed control of our government from its citizens.
Allowing no moral guidance, the globalist corporate entity will just expand until it has used up all resources on the planet, at which point it will collapse. Does Vinge's date of 2037 sound about right to you?
-- Ben Capoeman"
Normally I would link and not quote, but Jerry Pournelle has had his page for a long time, and it isn't set up like a blog. I hope this is not a problem to anyone.
Which also reminds me of a counter-singularity argument that Jaron Lanier made which can be summarised thus. "Machines are getting faster all the time, and theoretically more capable, but the software is shit" (He said more or less this in one of his articles on the subject.)
Red Rag to a Bull.
Let's see what's in this one:
1. Software is not self-designing. This is the essence of why Thingamy and SAP are different. SAP is designed to meet the majority requirements of the majority of organisations for the majority of the time. Because SAP has been doing this for upwards of thirty years, that's a lot of commodity processes that the software can support. It is also increasingly configurable, so that the amount of variation that you can make to the coverage is higher that it has been in the past. Nevertheless, there are still areas of active innovation, where the commodity process has not yet been invented. This is what Thingamy is good for, because it supports non-commodity thinking and execution.
2. Corporations are intended only to make profits, and in fact are punishable legally if they do not maximise this profit at all costs. Therefore, the letter writer argues, the acquired internal processes that are rewarded in corporations are those that maximise short term gains for profit, often by cost reduction and process streamlining. In some senses their feedback loops only head towards active financial acquisition, but may destroy other kinds of value.
3. Many kinds of value in the world are not noted by current economic thinking. I am thinking here of the current debate around post autistic economics, and some of the kinds of value that you and I can recognise, and that economics, at least in the neo-classical form cannot.
So, what's my so called point?
Well, if the processes that are available in the majority of organisations today are available as a commodity from SAP and others, and if the software design and build process is not going to create the AI type creation that is needed to drive the singularity, or whatever, then how do you create new businesses that can capture new value, especially that kind of value that classic economics ignores?
What for example if the real value of Google according to classic economics? The value of the machines in the server farm? The outstanding invoices for the ad-selling? Of course, the meanest analysis will tell you it is neither.
Hence, Thingamy, and kindred systems. Prototyping of news ways of doing business that might actually capture different value to the commodified process available in the current corporate landscape as captured by SAP and sold back to the market. (Now, I am not knocking commodification per se, for things that really add little value, it is the right thing to do.) Here is the market that I think that Thingamy is very useful for. Non-commodity process innovation.
It would be possible to replicate every function inside SAP in Thingamy eventually, but the problem with being a cheaper option for the existing commodity is that being cheapest is a very hard way to make a living. But if you could use Thingamy to generate a better business or product, then the sale of the organisation, or the value of the organisation's future cash-flow is in part based on the business model built into the Thingamy software.
Now, how Sig licences that value is a question for another time.
Finally, noticed how everyone is getting their knickers in a knot about patenting business methods? Now you know why, just one more barrier to the day that your business is commodified and zero profit, by building barriers to entry based on lawyers and not efficient execution.
Comments