News: 1638540135

  ARM Give a man a fire and he's warm for a day, but set fire to him and he's warm for the rest of his life (Terry Pratchett, Jingo)

When it comes to renting tech kit, things can get personal, very quickly

(2021/12/03)


Register Debate There's a reason why it's called a personal computer, and boy did this week's debate remind us all of the fact. We'd thrown open the motion that [1]Renting Hardware on a Subscription Basis is Bad for Customers .

The results were close. After all, traditional leasing has always been part of the corporate (non) ownership arsenal while the cloud already accounts for a vast amount of enterprise infrastructure.

This is how the readership voted:

[2]

JavaScript Disabled Please Enable JavaScript to use this feature.

[3]

[4]

The waters are further muddied by the fact that many workers will rarely be within spitting distance of an in-house techie for years to come.

So, why not hand over virtually all hardware to outside providers, who can bundle software and services, and deliver it all as a service. Shouldn't they do a better job of managing things, both on a pure asset management basis, and from an environmental point of view?

[5]

Veteran techie and Reg contributor Dominic Connor kicked things off. Could we expect straight talking and a flurry of comments? Dominic pointed out that yes, techies are having to, theoretically, support more home-based hardware, but that this is often kit that workers have provided themselves. Whoopee for employers, as he pointed out, "The cheapest computer is one you don't have to pay for."

Moreover, he said, companies may be wary of providers' "flexible" terms, having been caught out so often in the past.

More tangibly, he suggested, completely abdicating day to day hardware responsibility could lead to techies missing when big problems are quietly brewing. Renting infrastructure might make complete sense, he said, but renting user devices is still a solution in search of a problem. Vote no.

[6]

Dominic's arguments flushed out one pro-renting anonymous coward, who declared, "It's so much easier, who needs maintenance across a disparate workforce. We've got a vendor dealing with that headache, and it is truly the best thing I ever did."

How positive ... but they were swiftly downvoted.

As oiseau said, "The migraine (note the different malady) starts when you get shafted by a comma that you did not notice in the nth paragraph of the contract you signed or when your hardware landlord's operation goes titsup."

While Barry Rueger added, "Or when a company like Google arbitrarily shuts down an essential service on a whim. Contracts are fine, but ultimately it's about trust. Being able to launch a lawsuit isn't really a good replacement for having your business disrupted"

And trust didn't seem to extend to the folks in finance, with commenters recounting tales of "misunderstandings" on their part. Oiseau chipped in again with: "I have actually heard beancounters tell me that keeping a stock of DAT tapes to be able to do daily file server backups was an excess, that once every 72 hours was enough."

Anne Currie, another former CIO, weighed in on Tuesday, and said the answer is clear. The climate crisis means it's virtually certain that hiring will be a better option than owning – not just from an accounting point of view, but environmentally. This is certainly the case when it comes to infrastructure, she argued. When it comes to user devices, "Hardware management is non-trivial, and working devices too often have to be binned because of the end of security support."

So, surely it makes sense to have someone managing the lifecyle, who has "every incentive to maximize their asset, and are better placed than you to extend the life of it."

She closed with a warning, "Don't lock yourself into the old world and find yourself lumbered with unexpected, embodied carbon fees for your hardware when disposal time comes around. Don't think you will? Well, it's your funeral. Learn from 2020 – from now on, anything could happen."

iowe_iowe was heavily upvoted for wondering, "Maybe the question should be around what we're using vast exabytes of storage and teraflops of computing for. There must be loads of ROT (redundant, obsolete, trivial) data doing nothing other than occupying storage. Plus vast server farms dedicated to crypto currency mining, using the equivalent power consumption of a medium sized country. Plus big tech harvesting, storing and processing thousands of data points on every connected person on the planet, solely to serve up tailored adverts. Plus all the 'dark data' held by various govt agencies.."

While cyberdemon pointed out, "In order to calculate the environmental impact of 'what I do for my computing needs', I need to include the amount of power that Google, Amazon, Facebook, Microsoft et al spend rifling through the data I send them. If I don't use the cloud, then that rather large part of my personal carbon footprint becomes zero."

Let's just say many people were sceptical of the cloud's green credentials. And they didn't put much more faith in its security and privacy credentials either.

And, in a thoroughly relevant detour, wolfetone gave an excellent treatise on the shortcomings of the latest "eco" cars, and the implications for tech kit, "If you really, truly, cared about the environment - you would allow us to fix our equipment easily and at an acceptable cost. It's the one problem the world has that the cloud can't solve, and if it could, it would be the saviour of the environment."

While Jellied Eel speculated what the current lack of electricity due to storm damage would mean in our EV future, veteran tech and channel writer Billy Macinnes kicked off his argument with an admission: he uses a leased Macbook. The problem is, he continued, that vendors have a tendency to push ever more kit, more upgrades, whether they're really justified or not. So, until they stop this, renting hardware is not worthy of your full support, he said.

This left Neil Barnes in two minds – at least. "End-of-lease laptops are an excellent way for me to get a two- or three-year-old machine costing a grand or more for a couple of hundred quid. As the man said – if it's good enough, it's good enough. And when I *was* in the professional environment, it was rare that a case had to be made for other than more memory; we certainly didn't see machines dropping to bits with broken keyboards and screens and sockets after a couple of years." Yes, but they don't make things like they used to Neil.

Which is perhaps the point. msobkow added, "I purchased my phone on financing by setting up a 2 year phone contract. They claim it is 'free', but anyone with a functioning brain cell knows the vendor never loses on the deal; I'm being nickel and dimed (and then some) in various ways to pay for that 'free' phone. But thanks to Google's upgrade cycle, I may only get one to two years of 'owner' use out of it, due to their obsession with upgrades. My landline phone, when it was disconnected, was 23 years old..."

To which the obvious riposte is, if you'd carried on paying your bill you'd have an even older landline phone.

Thursday's contributor was me. The cloud has already won the argument when it comes to infrastructure, I argued, and with end user kit cheaper than ever, asset lifecycle management is, shall we say, lacking in many organisations. We've all seen the bulging cupboard of sparking laptops and disembowelled desktops at the heart of many tech teams. So, it makes perfect sense to hand it onto someone else and let the in-house techies do something far more useful. (Maybe even digital transformation.) How naive of me.

One commenter, sitt_europa, cited a previous example of an organisation that decided to focus on intellectual assets, rather than traditional assets. The organisation? Enron. Ouch. And an anonymous coward suggested companies looking to provide device as a service would be like crack dealers, with kit "always cheaper at the start to draw in new customers but rise with dependence.

In life sciences, the most expensive cost of running a genomics lab is the cloud service to process the data, and these costs are not coming down over time with the hardware." Very cynical. In fact, cynical was the operative word when it came to both sides of the argument.

But, cynicism, or at least scepticism, is precisely what is needed when it comes to evaluating subscription plans for hardware. They may well deliver benefits, but you're going to have to ask searching questions of any putative provider. And you're going to have to ask some pretty searching questions of your organisation to establish how efficiently – or not – you're handling things yourself.

What is unquestionable is that the issue really prompted some pretty fierce reactions. Both against the notion of device as a service, the motivations of anyone looking to provide such a service, the whole issue of tech and the climate ... and against our contributors' arguments. But, did the filth and the fury reflect the vote? Not exactly.

When it came down to it, readers voted against the motion by a margin of 4 per cent. Not exactly a landslide, but more than [7]certain referendums' results we've seen over the years. It seems there's a quiet majority out there who've decided they've got better things to do that look after hardware. ®

Get our [8]Tech Resources



[1] https://www.theregister.com/Debates/2021/11/29/renting_hardware/

[2] https://pubads.g.doubleclick.net/gampad/jump?co=1&iu=/6978/reg_offprem/paasiaas&sz=300x50%7C300x100%7C300x250%7C300x251%7C300x252%7C300x600%7C300x601&tile=2&c=2YaqhpOL@tfB5Uy488hEcfwAAAA8&t=ct%3Dns%26unitnum%3D2%26raptor%3Dcondor%26pos%3Dtop%26test%3D0

[3] https://pubads.g.doubleclick.net/gampad/jump?co=1&iu=/6978/reg_offprem/paasiaas&sz=300x50%7C300x100%7C300x250%7C300x251%7C300x252%7C300x600%7C300x601&tile=4&c=44YaqhpOL@tfB5Uy488hEcfwAAAA8&t=ct%3Dns%26unitnum%3D4%26raptor%3Dfalcon%26pos%3Dmid%26test%3D0

[4] https://pubads.g.doubleclick.net/gampad/jump?co=1&iu=/6978/reg_offprem/paasiaas&sz=300x50%7C300x100%7C300x250%7C300x251%7C300x252%7C300x600%7C300x601&tile=3&c=33YaqhpOL@tfB5Uy488hEcfwAAAA8&t=ct%3Dns%26unitnum%3D3%26raptor%3Deagle%26pos%3Dmid%26test%3D0

[5] https://pubads.g.doubleclick.net/gampad/jump?co=1&iu=/6978/reg_offprem/paasiaas&sz=300x50%7C300x100%7C300x250%7C300x251%7C300x252%7C300x600%7C300x601&tile=4&c=44YaqhpOL@tfB5Uy488hEcfwAAAA8&t=ct%3Dns%26unitnum%3D4%26raptor%3Dfalcon%26pos%3Dmid%26test%3D0

[6] https://pubads.g.doubleclick.net/gampad/jump?co=1&iu=/6978/reg_offprem/paasiaas&sz=300x50%7C300x100%7C300x250%7C300x251%7C300x252%7C300x600%7C300x601&tile=3&c=33YaqhpOL@tfB5Uy488hEcfwAAAA8&t=ct%3Dns%26unitnum%3D3%26raptor%3Deagle%26pos%3Dmid%26test%3D0

[7] https://www.theregister.com/2020/06/26/national_cream_tea_day/

[8] https://whitepapers.theregister.com/



The results speak for themselves

Snake

"But, cynicism, or at least scepticism, is precisely what is needed when it comes to evaluating subscription plans for hardware. They may well deliver benefits, but you're going to have to ask searching questions of any putative provider. And you're going to have to ask some pretty searching questions of your organisation to establish how efficiently – or not – you're handling things yourself."

Exactly. So the survey results exactly correspond to, what should be, the obvious answer to the survey's very question: "It depends" .

So, pretty much, a 50-50 split. Possibly yes, possibly no, it depends upon your individual situation.

Why wasn't THE major problem mentioned here?

nautica

I am completely surprised that The Register does NOT know how to formulate, and construct an unambiguous Proposition for the purpose have having a debate.

I am not surprised that addressing this lack of clarity and the vagueness---verging on equivocation, and double-entendre---and subsequent confusion on the part of the readers was studiously avoided in this 'results' recap.

If you're going to have a debate, Reg, do it correctly. That means: if you get ANY protests that the commenter can't understand YOUR rules of the debate, then you didn't do it right. You didn't, this time.

Re: Why wasn't THE major problem mentioned here?

JessicaRabbit

I find it quite a humorous coincidence that the vote is split exactly the same as the brexit referendum.

Re: Why wasn't THE major problem mentioned here?

doublelayer

This is very true. I have no doubt that there are people who think renting is a great way to handle every asset, but the comments alone make it clear it's not 52% of people. However, I can explain how the vote got that way. Here's a list of what not to do next time:

1. What hardware? Some people interpreted it as only servers, and therefore being a debate on cloud versus on prem. Since some articles mentioned desktops, this wasn't the only argument. This is a basic detail of the premise that the debaters didn't agree upon, so how could the readers know what the topic was?

2. What's the actual disagreement? Renting hardware is just bad in general? Because the debaters covered lots of different reasons, most of them completely ignored by others. A narrow debate like "Renting hardware is better for the environment" is clear, but a general one works well enough if you make the debaters talk about it in general. As it was, only Tuesday's article mentioned the environmental impact, and it didn't mention anything else, so it makes it seem like that debater was addressing a different proposal than everybody else.

3. One short statement on what the debate was. As it is now, the proposal is a paragraph of background that doesn't necessarily state an opinion. The debate proposal should be a simple, non-compound sentence that clearly states a point. Ideally, the sentence is so simple that the insertion or removal of a "not" clearly indicates what the against side thinks. In my opinion, you should also replace the for/against voting buttons with this sentence and its negation. In this case, the proposal could have been "Renting desktop equipment is a bad decision". The negation would say "Renting desktop equipment is NOT a bad decision". Unless it wasn't desktops, see point 1.

4. A little less necessary, but have the debaters respond to one another. We heard four speeches and now have to vote, but a lot of debates allow the debaters to make their addresses, then ask one another questions, then summarize what they think best supports their point. I think that's helpful.

5. Either allow people to change their votes or hold voting back until the arguments have ended. Again, this is more an opinion, but I think it helps make sure people are reading more than one article before casting a vote. When I vote on these debates, I tend to wait until Thursday to do it because sometimes the against side argues a point that I didn't think they would, so I no longer support them. If I disagree with Monday's article and vote against, that doesn't indicate that I support Tuesday's point.

Duhhhhh...

nautica

A short while ago, Nautica---being a closet statistician--noticed an interesting (possible) fact: given a large enough set of responses to a question, the results will almost always turn out to be 48 (or 49) to 52 (or 51). Percent, of course.

You say this ratio was close to that of "the Brexit referendum". Nautica ran his new-found 'finding' (sorry 'bout that) against several fairly large data sets, and then decided to check the largest data set on a binary question which is easily accessible to him: the last US presidential election's popular vote. Sure enough, if memory serves, Biden won the popular vote with something like 51% to Trump's 49%.

Nautica, being of the extremely modest type, decided to refer to this phenomenon as 'Nautica's Conjecture'. It is probably a well-known statistical phenomenon, possibly going by some well-known other name, but please don't burst his bubble; let Nautica wallow in his ill-informed bliss stupidity for just a little while longer.

We all would be most interested in hearing of your experiences or knowledge of similar results along these lines.

Re: Duhhhhh...

doublelayer

"then decided to check the largest data set on a binary question which is easily accessible to him: the last US presidential election's popular vote. Sure enough, if memory serves, Biden won the popular vote with something like 51% to Trump's 49%."

Yes, that election fits the pattern (51% to 47% including votes for other candidates), but it doesn't always hold. Just looking at U.S. presidential elections, there are several that don't end up going that way. 2008 was 53-46, 1996 was 49-41, 1984 was 59-41, and 1972 was 61-38. [1]Source . It's not surprising that there are many binary choices that don't have a clear majority, though. The reason is that a lot of the binary choices where there is a clear majority don't need to be asked (Do you like to eat food? Would you like to suffer an injury?). I'm not sure there's anything special about a 4% margin that makes it more likely.

[1] https://en.wikipedia.org/wiki/List_of_United_States_presidential_elections_by_popular_vote_margin

Re: Duhhhhh...

yetanotheraoc

"given a large enough set of responses to a question, the results will almost always turn out to be 48 (or 49) to 52 (or 51)."

Not "almost always", it's only true 52 percent of the time.

Unobfuscated Perl (#2)

A rogue group of Perl hackers has presented a plan to add a "use
really_goddamn_strict" pragma that would enforce readability and
UNobfuscation. With this pragma in force, the Perl compiler might say:

* Warning: Write-only code detected between lines 612 and 734. While this
code is perfectly legal, you won't have any clue what it does in two
weeks. I recommend you start over.

* Warning: Code at line 1,024 is indistinguishable from line noise or the
output of /dev/random

* Warning: Have you ever properly indented a piece of code in your entire
life? Evidently not.

* Warning: I think you can come up with a more descriptive variable name than
"foo" at line 1,523.

* Warning: Programmer attempting to re-invent the wheel at line 2,231.
There's a function that does the exact same thing on CPAN -- and it
actually works.