tag:blogger.com,1999:blog-6106782.comments2024-03-14T04:51:32.298+00:00Architecture, Data and IntelligenceRichard Veryardhttp://www.blogger.com/profile/04499123397533975655noreply@blogger.comBlogger436125tag:blogger.com,1999:blog-6106782.post-15794355643682440642021-08-12T22:20:00.978+01:002021-08-12T22:20:00.978+01:00The question whether we are mapping the data to th...The question whether we are mapping the data to the world or the world to the data is sometimes called Direction of Fit.<br /><br />https://plato.stanford.edu/entries/speech-acts/#DirFitRichard Veryardhttps://www.blogger.com/profile/04499123397533975655noreply@blogger.comtag:blogger.com,1999:blog-6106782.post-58332136508097379342021-08-11T08:55:26.427+01:002021-08-11T08:55:26.427+01:00But what you're talking about here, I think, i...But what you're talking about here, I think, is a metarational approach to data - which is a critical step forward.Benjamin Taylorhttps://www.blogger.com/profile/00513109479059624849noreply@blogger.comtag:blogger.com,1999:blog-6106782.post-61135460218861820622021-08-11T08:54:42.609+01:002021-08-11T08:54:42.609+01:00A bit of a side comment - but in the spirit of ...A bit of a side comment - but in the spirit of 'my electronic image in the machine may be more real than I am' (https://twitter.com/paulpangaro/status/1395385594402967554?s=20) and Shoshana Zuboff, of course one way to go is to ensure that those *are* self-fulfilling prophecies, give up trying to map to the world and map the world to your data. Surprisingly feasible if you have the power of (say) Facebook, and just what a game-theory AI would do!Benjamin Taylorhttps://www.blogger.com/profile/00513109479059624849noreply@blogger.comtag:blogger.com,1999:blog-6106782.post-70391169225633258422021-04-14T09:21:38.565+01:002021-04-14T09:21:38.565+01:00Thanks Scribe
Whether this counts as a software e...Thanks Scribe<br /><br />Whether this counts as a software error or something else is debatable, and I'm not convinced that improved software processes would have picked this up.<br /><br />In this particular case, it seems that the inaccuracy had some observable consequences (difference in the speed of the plane) but did not prevent the safe completion of the flight. What I do know about aviation safety is that near misses are taken a lot more seriously than in other forms of transport, which is why we are now able to read about this incident. (Atul Gawande has written about what healthcare has learned from aviation.)<br /><br />Meanwhile, I imagine that if the algorithm had more personal data instead of a simple (and possibly misleading) classification, up to and including counting how many drinks you had at the airport after checking in, it would be able to produce a much more accurate estimate, and this would allow greater fuel efficiency without compromising safety. There is always a plausible argument for greater surveillance, isn't there Scribe?Richard Veryardhttps://www.blogger.com/profile/04499123397533975655noreply@blogger.comtag:blogger.com,1999:blog-6106782.post-29637463503387126152021-04-13T20:38:44.431+01:002021-04-13T20:38:44.431+01:00Amazing - and yet not - how such a simple classifi...Amazing - and yet not - how such a simple classification/cultural mismatch can scale up. While the same issue _could_ affect individuals just as easily (eg medical software prescribes the wrong quantity of a drug), the fascinating thing here is a) how much the classification needs to scale up before it becomes a significant problem (eg light aircraft vs transatlantic airliner), and b) as it scales up, at what point does the "significance" demand a fundamentally different approach to risk management, and what funding/skills difference does that entail?<br /><br />Putting in manual tests and checks (like the pilot and the ground staff) is one thing, but who's to say 'the human touch' is more or less infallible than other check processes? Would heavily-specced software QA checks have caught this, or was the mis(s)-classification an inherent result of the company _needing_ to classify lighter passengers to save fuel?<br /><br />And if manually-driven testing processes aren't sufficient, would something like chaos testing be able to cover it? Generate random passenger data millions of times and then simulate the resulting flights and check if fuel levels and any other parameters are within acceptable levels.<br /><br />In this case, there doesn't seem to be much detail on what level of error is acceptable in the passenger modelling. Maybe the software error was actually acceptable in this case? (I know little of airplane safety...)<br />Scribehttps://www.blogger.com/profile/08757616056135886893noreply@blogger.comtag:blogger.com,1999:blog-6106782.post-87134041951000841262020-04-01T08:26:35.836+01:002020-04-01T08:26:35.836+01:00There is an important difference between data and ...There is an important difference between data and most other assets. A can continue to extract value from the data even after selling to B. A can also sell the same data to C and D. In your example of marine data, the question is whether exclusive access to the data has a significantly higher use value to one of the potential purchasers than non-exclusive access. In other words, is the data worth enough to Party 2 to pay you NOT to sell it to Party 1 and 3 as well?Richard Veryardhttps://www.blogger.com/profile/04499123397533975655noreply@blogger.comtag:blogger.com,1999:blog-6106782.post-46404359384671978162020-03-31T19:37:40.297+01:002020-03-31T19:37:40.297+01:00Thanks for your comments Iain
I completely agree ...Thanks for your comments Iain<br /><br />I completely agree that the use-value of an asset is context-dependent. If B can make better use of a given asset than A, in other words the use value for B is higher than the use value for A, then it makes sense for A to sell the asset to B, as long as the exchange price is anywhere between the two use values. The actual exchange price may depend on the relative negotiating skill of A and B, or on the presence of a market in which many similar assets are being exchanged.<br /><br />The question for B is this - if this asset is available for this price (e.g. market price), is it worth my purchasing the asset for this price. This can only properly be answered if B has some notion how much value they will be able to extract from the asset. As you say, this remains a fairly subjective judgement in most cases.Richard Veryardhttps://www.blogger.com/profile/04499123397533975655noreply@blogger.comtag:blogger.com,1999:blog-6106782.post-44404661424043567972020-03-25T17:17:57.177+00:002020-03-25T17:17:57.177+00:00An example of market value serves to illustrate th...An example of market value serves to illustrate the point. Let’s say you’re an international insurance company specialising in marine cargo, and you have a data set containing details of all marine vessels entering and leaving Rotterdam over the last 3 years, along with how many containers were reported missing for each vessel. You ‘sell’ the data set on a data marketplace such as Dawex. You might have 3 other parties interested.<br />• Party 1 is a an broking firm doing due diligence on a new client, the port operator. The data set is going to be worth a small amount as it will only be used as one of many pieces of information for a single deal. They might offer £100 for it.<br />• Party 2 is another insurance company that also specialises in marine cargo. For them, the data set has a high value because it can provide insight on a competitor. They might offer £5000 for it.<br />• Party 3 sells sensors for shipping container that link into an IoT network. For them the data is of medium value. They might offer a straight exchange for similar data they have about East Tilbury port and get their sales people to talk to the vendor in the hopes of selling the insurance company some sensors. <br />Similarly if you were calculating utility or replacement costs, you would probably arrive at different values if you were the CFO of the insurance company vs if you were a data scientist using the data to train an algorithm. <br />In short, I think there are too many variables involved to have a simple valuation process. It might be possible in years to come but it's a subjective exercise requiring experienced people unless it can be left to market forces.<br />Iain Heronhttps://www.blogger.com/profile/13647625954658212671noreply@blogger.comtag:blogger.com,1999:blog-6106782.post-61678068099401995052020-03-25T17:17:33.106+00:002020-03-25T17:17:33.106+00:00I still believe the best approach to managing data...I still believe the best approach to managing data as an asset is one based on the Asset Management standard, ISO 55000. This standard lays out the ‘what’ of asset management very clearly and can be used for intangible assets with a little work. The standard talks about maitenance of assets which needs a little interpretation for data, for example. The framework is designed to interoperate with other ISO frameworks including Quality Assurance and Information Security and adopts a business centred approach to data, which makes it ideal for asset management purposes.<br />Value measurements are difficult. You’ve described many methods above and I think there are several other in Doug Laneys book, Infornomics (apologies, I lent my copy out, so I can’t check). The method that I’ve had exec buy in for is contained in the HBR (https://hbr.org/2017/09/only-3-of-companies-data-meets-basic-quality-standards) but the key point here is that the value of any given data assets depends on the perspective of the person answering the question. Data is worth whatever a party is prepared to pay for it, and that holds true for any of the ways of putting a monetary value on it.<br />Iain Heronhttps://www.blogger.com/profile/13647625954658212671noreply@blogger.comtag:blogger.com,1999:blog-6106782.post-66926920240692099862020-02-10T17:04:28.505+00:002020-02-10T17:04:28.505+00:00Chris Argys and Peter Senge both have a lot of in...Chris Argys and Peter Senge both have a lot of interesting points to add in the discussion of double loop learning also. Doc Hollandhttps://www.blogger.com/profile/10542647581739173093noreply@blogger.comtag:blogger.com,1999:blog-6106782.post-51111010004159814972019-07-16T11:35:26.553+01:002019-07-16T11:35:26.553+01:00True, but academic ethics boards can sometimes be ...True, but academic ethics boards can sometimes be merely bureaucratic box-ticking. And Ben Goldacre talks about the <b>ethical paradox</b> in medical research, whereby the ethics committee blocks potentially useful initiatives. See my post <a href="https://demandingchange.blogspot.com/2019/04/ethics-committee-raises-alarm.html" rel="nofollow">Ethics committee raises alarm</a>. Dr Goldacre has also campaigned against about other ways experimental findings can be misrepresented and misused, including <b>publication bias</b>.Richard Veryardhttps://www.blogger.com/profile/04499123397533975655noreply@blogger.comtag:blogger.com,1999:blog-6106782.post-1374120653119181092019-07-15T18:57:58.874+01:002019-07-15T18:57:58.874+01:00The saying "Just an experiment" is so ou...The saying "Just an experiment" is so out of touch with the ethical development of scientific experiments that it's worrying - a lot of the scientific ethos (eg hypothesis testing for agile development) seems to have been imported by the private sector, but does this remark indicate that it's very much a "choose the favourable parts, ignore anything that is less favourable"?<br /><br />At a university, you would expect all experiments (especially ones with live participants) to be run past an ethics board and a procedure to consider harm. Obviously experiments from decades back lacked this insight - sadly, the results of these are only recently being questioned in many cases, and will take decades more to reverse in popular discourse outside of those discussions. Will the same be true of the tech we use in the second half of this century?<br />Grahamhttps://exmosis.netnoreply@blogger.comtag:blogger.com,1999:blog-6106782.post-32391252581980352872017-11-29T08:48:17.414+00:002017-11-29T08:48:17.414+00:00Well flat and matrix have similar characteristics,...Well flat and matrix have similar characteristics, yet flat is the best structure to used for responsibility distribution. Its used by tech companies because they want to keep things simple enough to be done by anybody. You can find more examples of flat and matrix organizational charts in the diagram community of Creately <a href="https://creately.com/Draw-Organizational-Charts-Online" rel="nofollow">Org Chart Software</a> .Evanhttps://www.blogger.com/profile/01134034541170679170noreply@blogger.comtag:blogger.com,1999:blog-6106782.post-64886452403501646742017-02-28T09:14:15.834+00:002017-02-28T09:14:15.834+00:00That's quite fascinating, for companies to sen...That's quite fascinating, for companies to send out emails that address you with your first name. You might not be impressed with <a href="http://nostosolutions.tumblr.com/post/156699022659/how-emojis-revolutionized-email-marketing" rel="nofollow">emails</a> like this, but I think it's quite effective. At least, brands are making efforts to communicate with you as an individual, rather than just another faceless customer. Lindseyhttps://www.blogger.com/profile/09279198171712692566noreply@blogger.comtag:blogger.com,1999:blog-6106782.post-48255301624014276362017-01-29T10:17:32.763+00:002017-01-29T10:17:32.763+00:00Interesting and well written post Richard. I'd...Interesting and well written post Richard. I'd love to know more about your "third order requirements" approach. Timhttps://www.blogger.com/profile/15189846953879912413noreply@blogger.comtag:blogger.com,1999:blog-6106782.post-87117619840257258972016-11-05T11:03:54.196+00:002016-11-05T11:03:54.196+00:00There are six basic building blocks that managers ...There are six basic building blocks that managers can use in constructing an organization which also known as elements of organizing or <a href="https://iedunote.com/organizational-structure-elements" rel="nofollow">organizational structure</a>.iHowPChttps://ihowpc.comnoreply@blogger.comtag:blogger.com,1999:blog-6106782.post-87602265823366337092015-04-22T09:47:40.780+01:002015-04-22T09:47:40.780+01:00Agile is a way to do more work relative to managin...Agile is a way to do more work relative to managing it, something like 80/20% rather than 20/80%.<br />Some agile is good for programmers. <br />Work management, planning is good for customers and executives to understand what, when... to expect, assess costs and prevent issues etc.<br /><br />In any case, the architecture of a system/enterprise is supposed to be delivered upfront, most of it in any case.<br />Adrian Grigoriunoreply@blogger.comtag:blogger.com,1999:blog-6106782.post-58907999746450592762014-11-09T22:08:22.501+00:002014-11-09T22:08:22.501+00:00Richard, actually just a quick comment on Business...Richard, actually just a quick comment on Business Scenarios.<br />They do indeed seem to be expressed in 9.1 as a solution requirements 'tool' (including technology requirements), which is at odds with the statement regarding their particular relevance in the initial Visioning phase. <br /><br />In this phase it does make sense to work through a series of Business Scenarios to help identify key business capabilities, however the 9.1 description of Business Scenarios would need to be changed to fit this kind of top down use, and is in my view misleading as it stands.<br /><br />Best wishes<br />RonRonShttps://www.blogger.com/profile/14007584315700448091noreply@blogger.comtag:blogger.com,1999:blog-6106782.post-40504258201644049372014-11-06T10:52:34.287+00:002014-11-06T10:52:34.287+00:00Adrian Grigoriu
The CMO-CIO connection gets a lot ...Adrian Grigoriu<br />The CMO-CIO connection gets a lot of attention lately. <br /><br />Unless the products and services are digital/IT, I see few reasons the CMO should have a special relationship with the CIO.<br />Yet, if the products and services are digital then IT is important because it is the core delivery technology for new products and features. Hence, in this case, the relationship should be indeed tightened.<br /><br />In addition, the social engagement technology is not provided by IT today.<br /><br />Anyway, it's nothing new about the fact that the IT and the business sides of an enterprise are, in general, disconnected.<br /><br />With advent of the cloud this connection would matter less because many internal and customer services would be provided by the cloud.<br /><br />Why is this relationship so important now?Unknownhttps://www.blogger.com/profile/18176782045304441420noreply@blogger.comtag:blogger.com,1999:blog-6106782.post-41116457952550995542014-11-02T19:48:02.316+00:002014-11-02T19:48:02.316+00:00Thank you very much Richard for remembering one of...Thank you very much Richard for remembering one of my basic principals when modeling any organisational patterns or ways of handling phenomens that one can observe in the world! <br /><br />My basic formulae of zero, one and many I also have adapted in a basic way of describing anything on three levels. The top level I refer to as the "conceptional level". It describes what you want, what is strived for, what must be underlying any other activities. You see the question word WHAT is anchored here. The middle level I call "infrastructional" because all questions based on HOW are answered with the methods contained in this level. The bottom level I refer to as "Physical" because here are the means anchored that are described by WITH WHAT. <br /><br />A small example concerning personal relationships:<br />Conceptionally I am convinced to have found my dream partner to live with. To work out how to realise that relationship I need to concern a lot of infrastructional aspects: Do we share the same appartement, do we share bed and briefcase, what to do with our respective children from previous relationships etc. On the bottom level we find each other attractive, share the bed, share the dinner table, occasionally live temporarily in each others appartement, share each others interests like books, musik, tv, politics, cultural and sports events and so forth.<br /><br />I hope this gives you an impression how the complete model that describes everything might be looking. As you see I'm still active in the thinking business.<br /><br />Hope to hear from you sometime,<br />your very old friend Raimo RikkiläRaimo Rikkilähttp://www.dffk-buxtehude.denoreply@blogger.comtag:blogger.com,1999:blog-6106782.post-10999114488295204122014-10-30T07:18:41.798+00:002014-10-30T07:18:41.798+00:00These methods all make some structural assumptions...These methods all make some structural assumptions about the dependencies between activities, capabilities, resources and other things. <a href="http://sistemieconsulenze.in-fad.net/" rel="nofollow">formazione fad roma</a>Elizabeth J. Nealhttps://www.blogger.com/profile/01824134730760179008noreply@blogger.comtag:blogger.com,1999:blog-6106782.post-61484788814858469982014-05-27T15:55:33.783+01:002014-05-27T15:55:33.783+01:00I can add a couple of recent papers:
1) “E-govern...I can add a couple of recent papers:<br /><br />1) “E-government reference model” (see http://www.samarin.biz/pubs/e-gov-reference-model-v1.pdf) <br />2) “Regional platform for e-governments and e-governance” (see http://www.samarin.biz/pubs/e-gov-platform-architecture-v2.pdf).<br /><br /><br />Thanks,<br />ASAlexhttps://www.blogger.com/profile/07571303538841911828noreply@blogger.comtag:blogger.com,1999:blog-6106782.post-14874665461682677042014-04-30T01:26:50.327+01:002014-04-30T01:26:50.327+01:00Taking into account the tiers of an EA, the busine...Taking into account the tiers of an EA, the business, technology and people. there is always the possibility, if not the reality, that the technology malserves the business operation in that the UIs are primitive, the processes poor, the data hard to get, the application too slow etc.<br />Alignment in this case means that technology can be properly adjusted, once one models the EA blueprint, to better serve the business operation. <br /><br />Another aspect is the difference in interests, skills, culture and aims between the business and IT departments, with IT seeming to have a mind of its own.<br /><br />While this relates to people organization it is still an EA issue, i.e. EA in the wider sense. Alignement often means that the IT has to be re-organised to serve better the business units rather than the grander IT intentions.<br /><br />In terms of different mindsets, the IT is often set to implement latest technology for instance while the business vows to keep any technology as long as it works reasonably cheap.<br />In term of skills and interests, for instance, people in ITwill easily move between industries, their skillsset being IT. So their interest is to keep updated their own skills with new technologies rather than work with the mainframes the business still employs.<br /><br />Misalignment maybe measured as a gap between the reality and proper business operation. Both business and IT architecture and the gap measurement and implementation are entities in the same EA space.<br /><br />But true, I've not seen yet how alignment can be shown on an EA (framework) structure except indeed in own GODS-FFLV. But this leads to a discussion about frameworks rather than alignment. <br /><br />To conclude, while EA uses today a lot of buzz words or metaphors if you like, some of them have roots in practice. Alignement does.<br /> Adrian Grigoriuhttp://www.ebizq.net/blogs/ea_matters/2014/04/business-models-are-not-strategies.phpnoreply@blogger.comtag:blogger.com,1999:blog-6106782.post-84455759596949441262013-12-22T10:31:19.275+00:002013-12-22T10:31:19.275+00:00Thanks Emeric
My experience is that organizations...Thanks Emeric<br /><br />My experience is that organizations never do exactly what their so-called "leaders" or "designers" intend, and that there is always some emergent behaviour.<br /><br />This behaviour may seem unnecessarily complicated or complex, but this is caused by events (or its interpretation of these events). A complex system "evolves" to achieve a locally efficient response to its perceived environment.<br /><br />Even if we regard an enterprise as an isolated system, as Emeric suggests, clumsy attempts to reduce this complexity are often counterproductive, merely shifting the system from its local optimum, and therefore making things worse rather than better. This is akin to what Deming calls tampering or meddling.<br /><br />Hence my view that the problem is usually not complexity as such but the failed attempts to deal with complexity.<br /><br />Emeric talks about people or organizations being comfortable with complexity. This is one of the benefits of (organizational) intelligence.Anonymoushttps://www.blogger.com/profile/11805047120034346106noreply@blogger.comtag:blogger.com,1999:blog-6106782.post-91913414835837309472013-12-21T14:53:21.698+00:002013-12-21T14:53:21.698+00:00I'm a bit late on this one... Felt between the...I'm a bit late on this one... Felt between the chairs. This time, I think that the "Anonymous" was me: @enectoux. <br /> Regarding "conservation of complexity", yes, basically this is the idea. Most of the time we are evolving in "an isolated system". E.g.: the Enterprise can be seen as an isolated system. Decreasing the complexity is either increasing it somewhere (which at the end doesn't help much) or it really decrease the other all system complexity, and then there is what we can call: "progress" or waste removal. <br />Even though, some of us are seeking for the 2nd alternative (removing some waste), the biggest part of the humanity is either comfortable in complexity (easy to blame) or try to give it to someone else (which is what I meant by "hide")<br />/EmericEmeric Nectouxhttps://www.blogger.com/profile/12155165729947265667noreply@blogger.com