Arm to drop up to 15 percent of staff, about 1k people(theregister.com) |
Arm to drop up to 15 percent of staff, about 1k people(theregister.com) |
This "cost-cutting exercise" (lovely euphemism) may not directly target technical people, but for some of them it may be a wake-up call nevertheless: "if they start lay-offs now, who says this will be the last round? And who says that I won't be next?". So they start looking around for alternatives, which I hear are plentiful at the moment, and hey presto, you got rid of the people you didn't actually want to get rid of...
So what's the thought process that goes into triggering the good people to leave? Is there just no other option?
And to add salt to the wound you will end up retaining those who are good at surviving such layoffs by doing some optical work. It's the worst of all the worlds.
It should also be pointed out that people are not plug-and-play parts of a machinery. They carry domain knowledge, context (Chesterton's fence), intra-company network they have painstakingly built and such intangible value. Besides, layoff doesn't mean the work disappears. It gets distributed, often unevenly, among existing workforce. I've seen it first hand and the aftereffects are just terrible.
That is, I've worked with people that didn't completely suck, but to quote Office Space, they always did the absolute bare minimum. Few companies have the Netflix mindset of letting people go with a large severance if they aren't top performers, so these mediocre folks can just be allowed to fester.
TBH, I've been glad as an engineer to see the company (belatedly) handle these sub-par performers, and to these folks benefit they get a fair severance and can usually find other jobs quickly.
Of course not all layoffs are like this, and often times layoffs at a large company are done "unevenly" as others have pointed out, but I have been in a company before where layoffs allowed the company to focus and perform better.
Every employee is either sub-par, at-par or above-par. If you perform the minimum to not get fired, you are necessarily doing what's expected of you, you're at-par (maybe you wont get promoted but you are doing what's expected of you), and if you are doing more than you need to then you are exceeding expectations.
This thinking that you can do what needs to be done to not get fired, but still be under performing just tells me its a bad manager(s) with no clear expectation setting and communicating it and if anyone is clearly sub-par making sure its communicated early with clear expectations instead of a doing a mass layoff to try and hope fix it.
https://computerhistory.org/blog/fairchild-and-the-fairchild...
There was a shock round of layoffs a few months ago, triggering a mass exodus. They’ve now had to start hiring again to make up numbers.
The amount of team disruption, lost domain knowledge and hiring expenses, especially as new people will want to be paid more, must have cost the business far more than they saved from the layoffs.
So short sighted.
Overall, cost-cutting in the short term. Just that.
It's shitty, cuz their playing with people's lives. As soon as the acquiring company takes over, their like "what's with all these unnecessary people?"
On the other hand, if you think the layoffs are due to an impending economic downturn, you could end up moving to a new job and being the new guy, so if they need to make cuts, you're first to go.
I was working for a small (~500 people) but well established startup around the time of the original dot-com bust. I have some friends that tried the same thing -- saw the writing on the wall and tried hopping to a bigger more established company for job protection, but they were the first to lose jobs when layoffs hit the industry. We had relatively layoffs in some areas of the company, but not in the core engineering team.
Far from worrisome, the talent may be fine with losing what they view (rightly or wrongly) as the extra weight.
A company in a bad downwards spiral is one thing, but this doesn't seem to be that. If there is a second layoff, maybe some fear would hit, but I don't think Arm is in a terrible spot. I wouldn't suggest that many people are thinking in dire terms.
I was in one large corporation (not large anymore) which did something like that. And sure, cafeteria was still full of marketing and sales people even after layoffs.
Not sure what went wrong, but people who have a clue have been bailing out left and right.
Reminds me of Amazon also - Something similar happening at AWS. When Bezos left, lot of people streaming out the door.
Steady talent exodus, and they are not the dumb people leaving. The great ones leave instantly. The trapped or mediocre (can’t leave…for some reason), too lazy, unmotivated etc stay and pay compounding interest and poison the company.
https://www.theregister.com/2022/02/08/arm_cancels_sale_to_n...
"Under the terms and conditions agreed by the companies, SoftBank Group will retain the fee of $1.25bn paid by Nvidia (this will be recorded as a profit in its fourth quarter) and Nvidia will keep hold of its 20-year Arm licence."
> As part of SoftBank’s deal to acquire Arm for £24bn in 2016, the UK Takeover Panel stipulated that it would have to double its UK headcount over a five-year period, which stood at 1,600 at the time. The requirement to do so expired last year.
"Some reason" is a reduction of operating expenses which boost profit margins. From an investor perspective it makes sense, ARM is an established brand that has limited headroom to grow, you can't expect hyper growth from them so they need to make a case that the ROI is worth it, especially now that a buyout is off the table.
Well, that "profit" comes at the expensive of some difficult to measure forms of capital. It is very likely a net-negative for the company, but it does look good on the books.
Also means any buyer has to cut less when the show up. Let the old mgmt take the blame for layoffs and culture changes, rebuild from scratch.
No you don't
>"Arm China, meanwhile, is contemplating a float of its own. The rogue Arm outpost, which is led by CEO Allen Wu and majority owned by Chinese entities, has become something of a problem for Arm because Wu was fired but has managed to retain his position."
Is this not a huge meatball that ARM has hanging out here? Further from a link in the linked article[1][2]
>"Following the investigation of Wu by Arm and Hopu, the Arm China board voted 7-1 to dismiss Wu. Given Wu’s refusal to vacate his role, Arm is growing anxious over the security of Arm China’s intellectual property, assets and finances, according to the people with knowledge of the matter."
How would any investor ever feel confident in investing in ARM unless this is resolved? Even if it is resolved what are the assurances that the IP is actually safe? This sounds like a debacle. Is this something that happened on Softbank's watch?
[1] https://www.scmp.com/tech/big-tech/article/3168443/arm-china...
[2] https://www.scmp.com/tech/big-tech/article/3089901/softbank-...
Yes, but mostly no. Most of ARM's market cap is tied up in IP; much of which is publicly available knowledge anyway. Wu stole said IP, but he has very few markets in which he's legally allowed to sell. It would be like someone setting up another Disney in China. Okay, so you can sell Mickey Mouse branded clothing in China, but you're definitely not going to be able to sell in any modern market.
ARM got off light. Millions of Western businesses have been destroyed this way.
It should be a stark warning for companies setting up shop in China: they will steal your technology, your IP, your facilities and assets, your trade secrets, your branding and trademarks, and all your equity. The government permits and encourages this. It's only a matter of time.
>"Wu stole said IP, but he has very few markets in which he's legally allowed to sell." Is this the consensus pretty much then?
Sounds good, but it is extremely likely a strong sign that the executive suite does not know what they are doing.
A company cannot achieve greatness by cost-cutting.
This is especially so in these times where it is so hard to find talent. Even if there are some redundancies, your key asset is an existing talent pool that has already been recruited and is up to speed on your products, technology, and internal processes. Strategically redeploying them to greater effect is a sign of knowledge & strength. Merely cutting and hacking is a sign of either incompetence, or that they DGAF and are trying to temporarily make the numbers look better for some near-term result (e.g., the likely pending new China IPO mentioned in the article), and leaving the resulting problems for their successors to deal with.
It is really rather obscene.
Big firings like this create bad blood, undermines loyalty to company and get many talented people to look at alternative companies to work.
A talented organization takes time to build up but is relatively quick to tear down. Skill in this industry is not exactly easy to get hold of. Talent which leaves may prove very hard to get back.
The person who can create a fab for less than a few million, while offering competing performance will be very rich.
Every bigger country will want some capacity to build chips.
It's possible to build something with next to no money, and I can't imagine it impractical to see some senior engineers gathering together and making a good funding pitch
They say so themselves: > "To stay competitive, we need to remove duplication of work now that we are one Arm; stop work that is no longer critical to our future success; and think about how we get work done."
I wonder if it's related to this?
In other words: we have no clue how to make use of a large force of capable workers.
That sounds like a company losing its grip. In this case to the likes of Apple and all the other companies designing closed off Arm derivatives.
Some of the talented ones leave instantly and maybe some of the great ones. But some really great ones stay because they are personally committed to the work and couldn't care less about their linked in or who's running the company. How much change was there at Bell Labs but the likes of Dennis Ritchie never left.
Talked about in this podcast: https://soundcloud.com/user-760920229/why-your-servers-suck-...
And an older article: https://oxide.computer/blog/compensation-as-a-reflection-of-...
People aren't leaving Amazon by following Bezos, people are leaving because of their terrible working conditions[0] and bad policies, including employees in corporate offices.
SiFive is a real competitor, but they are much smaller and offer only a fraction of the IP portfolio of ARM.
This is the simplest and most effective way to make Arm books look good, which is what regulators all around the world - including the UK - have decided is what's best for the world.
So this is not idiotic, this is effective, and is finance 101.
For Arm's long term success this is a very stupid decision. But that's not what any of the parties involved need or care about. What they care about is $$$ right now.
I won't be buying and holding Arm stock, particularly not as a long term investment, cause it looks like its going to be a poor one.
If indeed ARM needed to cut down on people, having a hiring freeze is the way to go.
It's not like Arm are facing long-term financial difficulties, their market share is presumably as good as ever, so getting rid of staff seems a bad idea.
The first resulted in every engineer in the team taking other jobs within weeks of notice. Quite a few were hired back at 3x their wage as a contractors as the projects while the talent found better pay and conditions and never returned. The company ended up paying more for its least influential engineers but without them it would have been in real trouble.
The second the mere threat caused 10% of the engineers to leave within weeks. Some stuck out the redundencies but clearly management had no clue who the talent was and ended up getting rid of some pretty important people. So the rest left gradually over the next year. The brain drain to competitors was very real, 15 years later they have not recovered and I receive regular calls to do contracts for them to bring some software engineering experience even a decade later, its a bit embarrassing really.
The third ended up getting rid of all the people who run a critical system in one of the UK's largest banks. I saw only the aftermath. All the software guys are long gone and its just full of system operators migrating binaries to more modern hardware, many of those systems are completely unmaintained now and have no team and no one knows where the code is stored, if it is still there. When the bugs started rolling in and management couldn't find a place to get them fixed they started hiring to replace the talent they had made redundent but all of them left within a week because the job was clearly impossible without source code which the company had lost.
I haven't yet seen a redundancy round result in anything but the almost complete exodus of talent and the gutting of a companies prior capability. I stuck around for one of them but I wouldn't do so again because the aftermath was horrible and all the people who made it interesting had left or were leaving. Everytime its been abundantly clear that who goes is mostly random, its not based on anything to do with importance to the company and the people that make the place tick know this.
[0] http://brucefwebster.com/2008/04/11/the-wetware-crisis-the-d...
If you offer a buyout to a bunch of people in a factory with no real alternative, you're selecting for the people who want to do nothing - which is probably the people who are doing (close to) nothing.
If you offer a buyout to a bunch of people with real alternatives, you're going to mostly buyout the people with the best alternatives (the most marketable employee / the most productive).
This is amplified by the fact that factory workers really have a minimum amount they must be productive. Whereas software engineers can LITERALLY do nothing. Further, factory workers have a pretty low limit on how productive they can be. The limit is much higher for software engineers.
It's an old practice that might have made sense, but definitely does not make sense in software.
ARM China now has the designs for several-year-old processors. Not that big of a deal.
- Instagram/Facebook
- Next/Apple
- Beats/Apple
- LucasArts/Pixar/Fox/Marvel (maybe)/Disney
- MetroPCS/Sprint/T-Mobile
my 2 cents is that those high-performers are not trusted enough with autonomy (by the increasingly large middle management). But the reason they are a high performer is because they're very good at working out what to do, as well as executing - which implies good autonomy, and self-reliant decision making. Perhaps at a large scale team, this sort of autonomy doesnt work.
As an anecdote, there are high performers in crafts such as blacksmithing - where the skills required are high and high autonomy is probably desired (i have never seen a blacksmith be micromanaged). So what makes software development so different that it cannot remain like a craft? My answer is that you have middle management who is made responsible for the output of the high performers, but they don't and can't truly control them - other than via micromanaging.
Agree. This is all anecdotal:
IMO - the people who get promoted to management in technical roles often don't have the people skills to perform the job well. I have had leadership thrust upon me multiple times in my career and it's very difficult. You're constantly balancing autonomy vs. active management for each person, in each role, for each unique situation. You're sussing out people's strengths and weaknesses working to help highly competent/technical people progress in their career. You're actively working to keep people on the same page, and sometimes you're working against your own contentious team members.
I rail hard against non-technical management managing a technical role - it just doesn't work. So, typically, technical people get thrown into management roles and I think we often suck at it. We don't take the time to work on ourselves to have the people skills required to be a positive leader of people. Then we compensate like any other crappy manager: we micromanage (crack the whip), we keep people in the dark ("mushroom management"), we DARVO, we have angry emotional outbursts, we pick favorites, etc.
What we see in software (and other technical roles) is that most people picked for management never put the work in to be an effective leader of people. And, I think the very nature of our work precludes many of us from developing the social skills it takes.
I think this is just an outcome of the reality that it's extremely rare for someone to be highly competent in people management, business, and tech. Such a person can work anywhere, and ask for any amount of money. Back in the real world, candidates will usually have certain strengths. Good companies attempt to train up said weaknesses, but oftentimes people are just not good at certain things.
I used to be on your side re non-tech managing tech, but I am now much more pragmatic. People management skills are very difficult, and I don't believe a people manager needs to be as technically skilled as their employees. Quite the opposite: the employees are there because of the skills they can bring to the team. The critical component here is ensuring the people manager is actually good. That means providing autonomy where appropriate, and structure where necessary.
i think instead of being picked, it's more that the management track is the only real way to gain higher compensation. There's a glass ceiling for IC (individual contributors) in terms of compensation - sure there will be the occasional principal developer (or some such like "architect"), but even then, those roles tend to turn into a managing role; you're herding other people and writing proposals and to convince upper management that such and such ideas for the product or system is good.
If such a person was just given the authority to dictate - or is working alone - they'd just _do_ it. They'd be able to accomplish far more in the same amount of time (provided they do have the skill and execution chops). But that's exactly what a high-performing person _is_!
Unfortunately, even if such a role was created and the person given the space, the compensation is not matching unless said person had equity holdings (such as being a technical founder). So to reach a level of compensation commensurate with such skills and capability, you end up needing to become part of the management - paradoxically removing the use of the very skill that would make you worthy of such compensation!
I contrast that with people that are laid off in a shrinking industry, and I have a ton of sympathy for those folks, because it's basically musical chairs around who gets to keep their jobs. Folks who are laid off in shrinking industries often need to take a large, permanent pay cut, or leave the workforce entirely.
How do you really equate your 'output' (value) with the money you are being paid? There's no direct 1:1 relation or easy formula. Even if you try to come up with an elaborate scheme of metrics, there really isn't a perfect way for it.
That said let's assume you could measure that somehow. Even then the underlying meaning of "bare minimum not to get fired" is not that you provide the exact 1:1 match of value you provide to the money you are getting paid (including benefits etc). There exists some X < 1, where you aren't getting fired but the value you provide is definitely smaller than the money you get paid. Is it 0.8:1? 0.5:1? 0.1:1?
Whatever that actual number is though, at most companies I've been at that number that lets you stay on is so low that nobody can argue that you're "just" doing your duty to extract as much value from the company for as little work as you can get away with (after all, in most cases the company is definitely trying to do that in reverse). It's so low that anybody that isn't doing the same thing gets severely demotivated. If I do my 1:1 or even a 0.8:1 or whatever and I like my job, then watching a 0.1:1 not being fired and even actively protected in some cases is awful.
New secure ARM Cortex's are launching everywhere too, so they are very much alive and well, for the next 20 years minimum.
For example, GigaDevices makes an STM32F1 clone which comes in two flavors with the same peripheral memory map: ARM Cortex-M3, or RV32IMAC.
Which big boys are designing their own CPU instead of buying from ARM? Apart from Apple and arguably Qualcomm with Nuvia?
And they will need to hold on to the ARM architecture and thus the license payments for quite the time... I bet Apple could go and develop a completely new ISA from scratch, but that would mean a third infrastructure switch in two decades on the PC and the first one on everything mobile - a lot of work for everyone involved with no real benefit, and especially after they spent multiple billions on getting their ARM stuff to the performance (in compute power and power efficiency) point they need.
Even the "biggest boys" aka nation states fail at making decent alternatives to the x86/64 and ARM duopoly - the Russian Elbrus and Indian SHAKTI CPUs are utter exotics and Chinese Loongson are a plain fork of MIPS (and again, completely exotic).
[1]: https://www.anandtech.com/show/7112/the-arm-diaries-part-1-h...
The Apple M1 lineup, for example. The M1 standard, pro, max, and ultra are not differentiated by their ARM cores other than the quantity of them. It's really the GPU cores and media encoding components. And that Neural Engine thing is really important for many apps and OS features. At what point do the general purpose CPU cores on a modern chip become minor players that have very little impact on overall chip performance? I don't think we're there yet, but it certainly could happen.
EDIT - I'm saying M1 vs M1 Pro vs M1 Max vs M1 Ultra. Not M1 vs Intel i7
[edit] Oh it says it right in the article: "SoftBank, Arm's owner, responded to that setback with a plan to float the chip design firm by March 2023."
You mean Europe. UK actually pays talent quite decent in comparison that's why many talents still flock to London from the mainland despite Brexit.
The rest of the UK... hit or miss.
Haven't deal with UK / EU a lot since COVID, so not sure if remote has changed that game up.
Large-scale layoffs very rarely (i.e. never) simply cut people that are "bad at [their jobs]."
I've also seen cases (and this appears more relevant in ARM's case) where the company just hired way too fast. It wasn't that people who were laid off were necessarily bad, but there were just way too many people for the amount of work, and so in absence of clear purpose a lot of people would have tons of "make work" meetings that were really there just so they didn't feel totally useless.
I'm certainly not saying all layoffs are like this, but I do take issue with the idea that layoffs never have a net positive effect on culture.
Regardless of if you are "[hiring] way too fast" or "finally [getting] rid of subpar performers," layoffs are still, by definition, firing a bunch of people's coworkers. Simple work efficiency, which is what you'd theoretically be improving in both of your examples, has relatively little to do with company culture, so your examples aren't very compelling evidence.
I've been a few times and I recall losing lots of friends and valuable team members. Even management typically doesn't spread the ridiculous myth that they're laying off people "who doesn't want to do their job and/or is bad at it".
Layoffs are heart-breaking and I can't earnestly believe that someone who has been through them would have this "counter-take".
Theres more than one way to skin a cat as they say, so layoffs can be seen in a few different lights.
Edit. Just because a company is listed on the stock market, like govt's doesnt mean they are not fronts for criminal activity!
Neither points to a very healthy culture to start with.
They spent the last 10 years beating out PICs, 8051s, etc. Because they were BETTER.
And now a million engineers now how to use them, and the tooling from 1000+ companies is all in place to support them.
They aren't going anywhere.
My experience writing C for RISC-V MCUs is about the same as doing so for Cortex-M platforms. They're a bit low-volume/expensive atm, but the core is very appealing to chip makers because it has solid software support and no royalties.
I also wouldn't write ARM off, though. Their 'coming-soon' MCU neural accelerators could open a lot of doors if it works well on a low power budget.
That seems to be said of a lot of things that do end up disappearing. Hard to tell whether there are real roots in the market or if it's going to evaporate like so many other things.
ARM will have to lower prices, as if you can save even just 1 cent per-device by using RISC-V in your microcontroller..