Disqus - Latest Comments for drewsherlock

Re: Climate change: the death of politics

drewsherlock — Sun, 06 Nov 2022 16:50:06 -0000

and vice versa?

Re: Immigration: the limit of tolerance

drewsherlock — Thu, 03 Nov 2022 04:17:17 -0000

"In the space of ten years, therefore, nearly a million people from India have taken up residence in this overcrowded country".

I don't follow this. If in 2011 the figure was 694,148 for Indian born residents and in 2021 it was 920,361, didn't 226,00 arrive (plus however many of the original residents have since died)? Not a million.

Re: Politics: a person of colour

drewsherlock — Tue, 25 Oct 2022 06:05:06 -0000

Seems to me his skin colour and religion are the least defining aspects of his background.

Each of the last 5 PMs went to Oxford. 3 went to public schools. We're not drawing from a very deep pool.

Sunak is just more of the same, with a different skin tone.

Re: Politics: madder than a mad hatter’s tea party

drewsherlock — Mon, 26 Sep 2022 07:52:47 -0000

Well of course.

Re: Politics: madder than a mad hatter’s tea party

drewsherlock — Mon, 26 Sep 2022 07:21:22 -0000

It's 20 GW of power derived from some form of storage.

Re: Politics: madder than a mad hatter’s tea party

drewsherlock — Mon, 26 Sep 2022 03:45:19 -0000

They refer to 20 GW of storage in the context of balancing the grid, so are talking about the power available at a point in time (GW) not total energy available (GWh).

Re: Climate change: on being prepared

drewsherlock — Sun, 18 Jul 2021 14:53:17 -0000

Comparing energy costs fairly is tricky, but my understanding is that we're pretty much there already:

https://www.forbes.com/site...

Re: Climate change: on being prepared

drewsherlock — Sun, 18 Jul 2021 14:37:01 -0000

I don't opine on the worth of climate models as it's well outside of my area. That's my entire point.

Re: Climate change: on being prepared

drewsherlock — Sun, 18 Jul 2021 12:04:08 -0000

Again yes. I've no problem with a lively debate on the implications on policy of the current range of scientific opinions. And of course, the science will be constantly move as measurements and models advance.

I object to denigrating the science because its political implications are unpalatable.

There seems to be little serious science that refutes that anthropogenic climate change is happening and may have significant societal impacts. If one wants to dispute the science one needs to do better than some guy on the internet says greenhouse gases are a hoax, which is where this exchange started.

Re: Climate change: on being prepared

drewsherlock — Sun, 18 Jul 2021 11:29:06 -0000

Well yes - climate science is complex. Maybe we should trust climate scientists to gather evidence, build models and argue over all this to come to a consensus opinion, rather than base our opinions on science illiterate nonsense on the web.

Re: Climate change: on being prepared

drewsherlock — Sun, 18 Jul 2021 11:00:10 -0000

That link is, to use a technical term, bollocks.

He completely ignores the key fact that that the absorption rate of the greenhouse gases varies with the wavelength of the radiation. The gases are pretty much transparent to incoming solar radiation, but somewhat absorptive to longer wavelength infrared radiation from the earth's surface which it then re-emits in all directions, impeding energy flow from the earth's surface back into space.

With or without an atmosphere with greenhouse gases, the earth is essentially in energy balance. Without an atmosphere, the black body temperature of the earth to emit the same energy as is incoming from the sun is -18C. Given the impedance the greenhouse gases put on outgoing radiation, the surface temperature needs to rise in order to come back into balance.

At present an average surface temperature of about 15C brings this into balance. With increased greenhouse gases this will rise.

Re: Pete North Politics Blog: Shifting battlegrounds

drewsherlock — Sun, 01 Dec 2019 10:04:24 -0000

"There was a time when producing goods to meet different regulatory specifications required two production lines and different print runs for country specific labelling. That was a serious problem before the advent of cheap industrial laser printing and advanced bespoke manufacturing techniques. It's even less of an issue now that technicalstandards have gone global. These days producing to a different standard just means selecting a different mode on the operating software."

This may be true for some industries, but I'm not sure it generalises to all or even most.

Re: Brexit: the fringes of madness

drewsherlock — Tue, 03 Sep 2019 12:19:29 -0000

but to get a general election doesn't he have to lose a VONC (assuming he can't get 2/3 of MPs to vote for an election)? Labour could either not call for a VONC (can the government call a VONC in itself?), or support the government citing not being willing to fall into the govt's trap.

He'd be still in power with legislation to compel him to seek an extension.

Am I misunderstanding?

Re: Brexit: ours is but to do or [panic] buy

drewsherlock — Wed, 21 Aug 2019 12:20:41 -0000

Panic buying is a wise strategy if everyone else is.

Re: Brexit: ours is but to do or [panic] buy

drewsherlock — Wed, 21 Aug 2019 11:47:56 -0000

A quick Google search (https://www.euronews.com/20... finds the CBI estimates average tariff on UK exports to the EU would be 4.3% (average on imports from the EU into the UK would be around 5.7%, but we're setting that to 0%!)

So, yes, if we did this for all industries, we'd be subsidising them by around a quarter more.

Re: Brexit: ours is but to do or [panic] buy

drewsherlock — Wed, 21 Aug 2019 11:36:25 -0000

Back of fag packet calc:

UK export to EU in 2018: £289bn (https://researchbriefings.p...

UK net contribution to EU 2018: £8.9bn (https://fullfact.org/europe...

8.9 / 289 = 3.1%

It's probably order of magnitude similar.

Re: Brexit: ours is but to do or [panic] buy

drewsherlock — Wed, 21 Aug 2019 08:45:14 -0000

It adds up though if you do this for every industry where there's going to be tariffs to export to EU.

Re: Brexit: living in hope

drewsherlock — Sat, 04 May 2019 11:20:50 -0000

30,000 divided by 8,000 seats is about 4 per seat.

Re: Brexit: and finally …

drewsherlock — Fri, 22 Mar 2019 11:55:31 -0000

3 years ago one of the nutters murdered one of their number. You can see why they might be jumpy.

Re: Brexit: red tape blues

drewsherlock — Fri, 25 Jan 2019 10:54:50 -0000

I don't buy that in the medium term. Comac hired 2,500 of China's best graduate engineers. Go around any UK university engineering department and you'll see maybe 20% Chinese students. Digital manufacturing, composites, additive manufacturing - China can as easily buy or develop these technologies as anyone.

Re: Brexit: red tape blues

drewsherlock — Fri, 25 Jan 2019 10:25:08 -0000

The first of Comac's C919 narrow body airliner is built and flying. Granted it's probably at least half a generation behind Airbus and Boeing, but they are investing heavily and, as you say, they have their own market whose growth will be significantly higher than the West's. Having spent some time at their R&D facility in Shanghai, I wouldn't bet against them in the medium term.

Re: A Maturity Model for Product Data Accessibility?

drewsherlock — Wed, 14 Dec 2011 17:48:37 -0000

Jim, it's great you got this conversation started.

The range of replies, I think, demonstrates that there are numerous tasks out there where the lack of appropriate tools to access and retrieve product data inhibits good decisions and process improvement - not just in engineering, but also (perhaps more so) in 'out-of-engineering' functions.

The key observation is “accessing product data and centralizing it are not absolutely linked.” The 'management' bit of PDM or PLM is vital in many instances (but not all), however it tends to be 'heavy' to set up and run. Indeed it may not be immediately possible as in the acquisition example. Once we recognise that product data access and retrieval is separate (though complementary) to product data 'management, then we can look for new technologies, most likely from other domains such as the web or semantic search, to solve these problems.

What those technologies are, and what the use-case sweet spots are, is up in the air at the moment. That's what has made this discussion so interesting...

Re: Accessing All of Your Product Data Regardless of Where and How it is Stored

drewsherlock — Tue, 06 Dec 2011 19:16:06 -0000

Posted this on LinkedIn, but thought I'd put it here as well....

Jim, interesting paper. Got lots of thoughts mostly echoing yours, but a few of my own as well ;-)

Retrieval tools are poor

Regardless of whether you have PDM/PLM or not, the current tools we have for retrieving data from stores of product data are pretty poor. Things are OK when you want to retrieve items in 'standard' ways through part names, descriptions or attributes or by browsing through product structure. But try to get your system to answer an engineering question like 'which of our stainless steel parts are overpriced - show me all stainless parts ranked by the ratio of price to volume'. The data's there in your PDM and ERP sytems but can you get at it that way?

There's almost never only a single source of information or single type of user.

Over the last few months I have spent several days sitting next to engineers, and others outside of engineering, being shown how they get at the information they need. It's fascinating watching them wade through multiple documents (often not electronic), then dive into CAD assemblies to find a particular part, search through PDM systems, cutting and pasting part numbers into ERP systems all in order to get at a particular piece of information. It's real swivel chair integration.

Of course, the nature of this journey through the data varies according to what data managament systems you have in place, but there is always a boundary to your PDM/PLM system and some data thats's not in it. Either it's unmanaged or handled in ERP, SCM or CRM sytems. As Chad Jackson points out on his blog, http://bit.ly/v4eUl1, engineers have a very broad reach across all the departments and systems in an organisation.

There's also not a single type of user. There's often real benefits to be had by having users in purchasing departments or sales departments, say, be able to access product data, but maybe the PLM and CAD systems don't reach there because of licensing costs or skill sets of those users. And so these users either have a much harder journey navigating through the data, or they don't bother and bad decisions are made.

Retrieve or manage or both

I have seen a number of companies whose product data (CAD files, document, spreadsheets) has grown quite large and it gets hard to find stuff. PDM is offered as a solution and PDM and/or PLM does a good job at 'managing' the data, revision control, ECOs, other workflows etc.. However, in at least some cases, it's probably not worth the effort and expense to migrate data into the system. But, as you point out, you don't necessarily have to centralize data in order to make it findable - just crawl and index the data and provide good retrieval tools and it may be good enough.

Scope for some interesting innovation

You've mentioned a couple of emerging solutions based on existing search technologies (semantic or not). I think these can take you good distance, but I'd argue that the nature of product data is different to the document and textual data most search is oriented towards. There's more structure and large amounts of data in CAD files describing the geometry of the product to an incredible level of detail which is opaque to these techniques.

Similarly, designing the UI specifically to display product data (especially where this is associated with 3D objects) can really enhance the effectiveness of the software. Dassualt and Siemens have done some interesting things here with 3D Live and HD-PLM.

Now the plug...

We at ShapeSpace have been working with Actify on the search in their new Centro product. Basically, Centro is intended as a way to do 3D product data mashups, pulling data from multiple sources and allowing it to be aggregated, searched and results displayed in a 3D context. See http://bit.ly/vYvwj8 and http://bit.ly/rVPJsu

Nice thought provoking paper! I think there's some interesting new ideas to come out of this area in the future.

Re: Deriving form from function

drewsherlock — Tue, 13 Sep 2011 11:04:47 -0000

Really interesting, some great stuff comes out of the Spatial Automation Lab.

I know you were teasing us, but your title oversells it a bit. I don't think it's a "general solution to deriving form from function".

I think computers are good for exploring a space of possibilities (see algorithms for topology optimization, shape optimization and now for exploring configuration space). They still need a human to give a machine understandable definition of the function required and a design space through which to search. They can then be used as a tool to help identify a form to meet the required function. It's a few steps less than 'automatically' deriving form from function.

However, it'd be very powerful to have this buried in your CAD system, just as having a constraint solver or a modeling kernel are.

Re: Cloud PLM and Amazon S3 Scale

drewsherlock — Tue, 26 Jul 2011 11:19:21 -0000

It's interesting to think that there have been around 4 billion parts, including fasteners, generated for the 747-400s alone. Gives some perspective over the number of parts overall knocking about, especially in the context of some of the ideas around the Web of Things.

Not sure the issues with product data are to do with scale as such, more to do with the issues you mention: granularity, redundancy, latency, bandwidth. Deciding where to store data so that it's 'near' the user. And for CAD, do you move pixels around, or triangles or b-rep.