These are likely seasonal job losses like retail or related. You have to do Year over Year comparisons for jobs. Month over month comparisons are misleading because of seasonality
If January had the highest number of jobs since the Great Depression and we have returned to the March mean, this could still be true while there was absolutely nothing interesting going on.
If there was an unusually high peak of seasonal jobs just prior to the months in question, that would be an utterly meaningless statistic. If it's a moderate random drop coinciding with the end of a somewhat high seasonal number, it's still not relevant to long-term outlook. That's why monthly numbers are difficult to use in isolation.
You need to read the whole article to get to the statement:
“Unemployment rates throughout the Bay Area are among the lowest in the state. Almost anywhere else in the country, the Bay Area’s job trends would be the envy of that region.”
Every downturn has to start somewhere, and maybe this is the beginning of a bad stretch, but the largest employers are doing fine based on stock prices. Apple, Alphabet, Cisco, Facebook all made new all-time highs within the last month (or week).
Here is the March 2017 unemployment report. 3.5% for Santa Clara County. And states that jobs are + 2,000 from January to February.
Stock price does tend to be a leading indicator, because the major Wall Street analysts all do their best to independently derive a company's sales so they can front-run the stock before official earnings come out. Causality tends to run:
A number of my college classmates went into equity analysis or management consulting positions, they would do things like canvas retail locations and count the number of shoppers walking out with the products of interest, or track search engine rankings, or sit on Google Trends to watch for changes in search volumes in the markets of interest.
This is also why stock prices fall when a company "misses estimates", even if earnings are themselves up. Wall Street has a consensus estimate for how they expect the stock to do, and that estimate is priced into the stock. When the analysts are wrong (which isn't uncommon), the stock corrects based on the difference between the estimate and the reported earnings.
Sure, it's never perfect, but in context here it's relevant. Grandparent's point is that there is no evidence of a broad-based downturn in Silicon Valley, based on stock prices of major tech companies. Yours is that stock prices tend to be lagging indicators. I don't believe this is true in the general case (certainly it can be in specific ones), but if you believe that it is - how big is your short? If you believe that stock prices are lagging indicators and market consensus doesn't reflect reality, this is your chance to make money off that belief.
With this logic, most of the bursts wouldnt happen, would they? In 2000 with the dotcom, why was it a bust and not a gradual decline? Same goes for 2007/8 bust. If it were not lagging, wouldnt you expect people would have realized this and bring market to its stable levels with similar shorting?
When you say "in the east bay" that's confusing to me. Are you only considering companies with an East Bay office? I live in the East Bay and don't make any secret of it, but I get recruiters contacting me from all over the SF Bay Area (and outside of it too, for that matter). Would you not work in SF if someone wanted to hire you there?
> “Right now it’s tough to find a job in I.T.,” said Steve Satariano, a San Jose resident who has been working in contracted, temporary information technology jobs in recent years. “I’d prefer to find a full-time, permanent job, but in I.T., beggars can’t be choosers. It’s a little frustrating because tech companies always want to try before they buy.”
The numbers don't really back the first part of the statement up...
There are +3000 more I.T. jobs from 2016->2017 and I.T. only lost -500 jobs since last month, but that is a single month while the numbers are still higher than last year (down 0.7% monthly but up 4.1% overall yearly).
And "Computer Systems Design & Related Services" also grew 3.7% year over year with 0% loss in the last month.
The news is always looking for the negative 'sky is falling' angles and they will always be able to find someone to provide a sound bites to back up the story. But as usual the numbers say otherwise or are simply neutral/boring.
Some people are obsessed with being the first one to call the new demise of tech, or the latest bubble. But if anything the industry has become more resilient. Which is what happens when industries mature and grow larger. With more diversity and a larger base of employers/employees then fluctuations are naturally going to be smaller and better contained. That plus greater domain expertise develops over time - companies/investors learn how to manage risk better and understand the nature of the marketplace so overall losses are smaller.
Correct - if tech is still getting H1B visas, that means there are jobs that the companies claim were unable to be filled by US citizens. They have to go so far as to post two conspicuous notices of the application for visas in their workplaces [1].
When people end up training their replacement H1B worker (ask the Disney employees for one example), I don't think companies are following the spirit or the letter of the rules. The ease of skirting the issue shows up anytime someone looks at the top 10 H1B companies.
There is a significant difference within different roles in I.T. As someone who's worked in the cost center side of the industry, I can relate with the author. It's a problem not limited to Oakland or San Jose. Helpdesk, Desktop Support, and related jobs are commonly temp to perm with horrible pay for the amount of stress.
For sure. As a sysadmin/network admin/IT manager, it's a buyers market. As a DevOps/infrastructure engineer, it's a seller's market.
I think training is part of the problem. More importantly, workers need consulting help around how to take existing skills and use them to apply to upmarket jobs with higher pay.
Since you brought up IQ, I'd guess this was something about the bell curve. It's not obvious what "nuance" could have to do with that, however. Perhaps that was an unhinged spell-correction of "variance"? Sometimes I wonder if people post with imprecision because they think that way...
Richard Herrnstein certainly never suggested that there was a chunky or sharp delineation built into the bell curve, It's probably fair to say the manifestation of capability and capacity of the human throughout the spectrum specifically in the context of the article would likely require to you see no "gradation".
It's weird that people who claim unpopular opinions, often at the expense of others, so rarely can find the time to argue their excellent ideas. I have a number of friends that works in "IT" in Europe. The difference between working for a US and European "IT" company is night and day. It simply has very little status as a field in the US. If you had ever seen the things a systems administrators have to deal with from highly educated engineers, you would know that it has less to do with "IQ" and everything to do with domain knowledge.
I've had the pleasure of working with some great a Linux sysadmins. I had a great relationship with them as a developer. Another developer hated them and treated them poorly. Do you think they would help him out after hours if they weren't on call? Same thing for EEs who treat technicians poorly. A tech might tell them a rework is impossible. If they like you and you buy them a beer from time to time, they might say, "That's going to be really tough, but I'll give it a shot and we'll see what happens."
I think a great piece of advice for life is, "Never assume you are the smartest person in the room." Remember that everyone brings something to the table.
I agree. The sysadmin people I have worked with in Europe in have generally been very clever people that I loved to have on my team. The IT/sysadmin people I experienced in Silicon Valley in my next company though.. meh.
Not that I agree with the specific opinion in question, but when it comes to cause-effect questions with a moral angle, you're going to convince literally nobody by arguing. I never take the time, because it doesn't bother me in the slightest whether others agree with me, or are wrong.
I wouldn't call Ansible and Terraform programming languages. In two and a half years at a well known startup, I've written less than 1000 lines of code in a senior DevOps/Infra role.
DevOps/infrastructure is knowing how to glue together off the shelf tools and understanding how everything works. Great that as a dev you can probably bolt these tools together with docs, but your lack of context with the underlying fundamentals will eventually bite you.
This is dead but I thought I might comment anyway..
Your comments convey a very narrow perspective of the job market that falls under "DevOps". That is to say, the sphere of roles looking to be filled by companies who believe they are filling a "DevOps" role is much larger than what you believe. It almost seems as if you are limiting yourself and projecting that onto the entire job market.
Now, I'm not sure you consider your current position as "limiting".. However, I can say that there are MANY roles being filled for "DevOps" candidates that extend beyond writing Ansible roles or Chef cookbooks. Speaking as someone 6 years into their engineering career and as the person who wrote the initial powershell(as a linux cloud engineer!) feature provider for Chef, I would not have taken a role such as you describe after year 1.
Now, you may be in a position where writing Ansible or Chef resources long term sounds swell. That's fare! But there are many, MANY roles that fall into the systems engineer/SRE bucket that are labeled at DevOps! Honestly, they are a matter of degree! The higher-end roles are senior, lead, unique, and require heavy development skills. If it's not the provisioning platform you are developing and supporting, it's the performance characteristics of the product itself.
> DevOps/infrastructure is knowing how to glue together off the shelf tools and understanding how everything works.
That's actually not that much removed from quite a few programming jobs. It's mostly different manuals and the occasional plug to fit into a socket but other than that the differences are smaller than the similarities and debugging skills are important for both of them.
So what were sysadmins before the term DevOps was invented ~5y ago?
Unless you are arguing that people never did large scale automation
of processes around deploying systems and software until this term existed...
In my personal vocabulary - there were always 2 kinds of sysadmins:
- Real sysadmins- which is now the term 'DevOps'
- IT System Technicians - which is what sysadmin is being perverted into
The old-style breakdown typically depended on how systems/development minded the company was - with more forward looking companies seeing their sysadmins as the former category and people who "administered" e.g. 'tended to' the IT "system", and the reactive companies who viewed IT as some foreign expense to be managed viewing their sysadmins as advanced IT techs who were 'admins' (in the administrative assistant sense) for the various 'systems' (e.g. computers)
that they purchased.
As someone who always operated in the mindset of the former category,
but unfortunately has operated mostly in the latter type of company,
it is difficult to make the transition into 'DevOps' roles since this term
has been subsequently invented to describe the work that I always did,
using tools that I would have written from scratch myself in the old days,
or deployed myself in the current days, but unfortunately due
to crappy management and backwards modes of thinking, don't have the proper
'current' job-title on my resume (e.g. sysadmin)
No one is going to check that your title matched at most companies I've worked. Yes if you list yourself as VP of Finance and you were actually an Accounts Payable administrator, you will have some trouble. But that trouble will start far before the background check. If you're in a DevOps role that happens to be called System Administrator, simply change what you call yourself. You will still have to interview as a DevOps person of course -- titles are not reality -- but you'll get far more inbound interest, guaranteed. And according to what you've written here, you won't be lying or even misleading the people who want to talk to you.
If you talk to someone about their day and it revolves around tickets, they are the techs that you're talking about. The real "sysadmins"/devops understand systems and own things.
It's ok to realign your title with function. Every company handles this sort of thing differently. I was a "Computer Systems Programmer 3" for two years. In that organization, that rank was roughly equivalent to a Staff Engineer. My resume says "Lead DBA", as that was my primary function... the formal title assigned was just whatever was available when I was promoted... the title description was written for mainframe people!
> workers need consulting help around how to take existing skills and use them to apply to upmarket jobs with higher pay.
> DevOps/infrastructure engineer is basically a programming position
sysadmin/network admin ==>> DevOps
IT manager => Manager or lead.
DevOps is not a programming position at all. It is usually cheap system administration (that replace the term sysadmin) and it does NOT require coding. Pay attention to the name and the place you joins, you're can easily be burnt by joining a team of operations monkeys while thinking they were the dev and system type persons.
Exactly, the role of the 'IT person' as the system administrator and keeper of updates on things like the Novell server, has given way to Google Apps and Microsoft's Office 365. Small businesses who could barely afford to pay someone before can now eliminate that role and still have email, shared files, back end inventory management and accounts management, and the occasional web developer contract to keep the web site up to date.
Personal services IT is still alive and well and a number of people I know have made a living out of helping older people manage passwords, upgrade their systems, move their phone plans or transfer their data when they get a new phone etc. But to be successful at that you also need to talk to people and be able to maintain a business relationship with them, not a skill that everyone has in addition to their deep knowledge of IT.
As tm2d mentions devops is still a hot job market. But it is not the small business 'tech' role so there are fewer actual slots for that role.
There was also a comment in the article about H1-B visas and employers wanting "younger and less expensive" workers. I try to remind my older friends that if someone can spend 6 months learning to do what you do and do it well enough to meet the needs of the job, then you are only "worth" what a company would pay that person they just hired. If you want to have a larger salary and better job security, then you need to be able to do things that can't be trained in 6 months. The days of 'too few programmers to go around' are long past, there is now a surplus and they are coming fast and furious out of college.
The last time I listed a $120K job opening I had hundreds of resumes thrown at me, over 10[1] of them were completely qualified and So perhaps it would be more accurate to say "$80K/yr programmers are scarce."
In 1999 when I was hiring programmers and offering above market starting salaries I wouldn't get any qualified resumes.
It is from that experience that would assert it is a 'pricing' issue rather than a 'selection' issue.
[1] After 10 qualified I had my first interview round, 3 were brought back for secondary interviews and one hired. I expect there were easily 20 - 30 engineers in there that could have done the job.
Mine too. We're having real problems filling contract positions. People are juggling multiple offers, and if we don't jump on a candidate right away they take another offer and we have to restart the process.
We've gone through surplus years before, and they don't look like this.
I think there's a misconception about performance some perception issues about the ability to do the job. In 6 months of training you might get someone who can sort of do the job, but probably not very well and not at the level of a "real" professional. A good example of this is code camps. They churn out a ton of programmers but very few people that are actual engineers or can carry out actual projects.
But to management/high ups, they "can do the job" and are cheaper. It's like outsourcing. You save in costs but there are hidden downsides and risks.
I'd probably argue that very few engineer jobs can be done competently with a few months of training.
This was true even 14 years ago, especially after the dot-com bust when many former developers took jobs as help desk techs when the job market fell apart. I was one of them.
I eventually figured out there were very limited opportunities for someone with my skills and interests in that role so I studied and worked to build the skills to get a software engineering job.
I think that Helpdesk roles in general should be seen as stepping stones toward careers in more advanced roles. I actually think my experience doing that work has made me a better engineer and coworker.
Helpdesk was my stepping stone as a degree less high school dropout getting into a software development role. They can certainly serve a purpose, but I wouldn't stay there longer than I had too given a choice.
There are still a "significant number" of I.T. tech job openings that finding a job should not be tough, but "getting" those job is another matter. Poor interviewing skills may make it more difficult to get jobs, but "try before you buy" is how I got my job over 10 years ago and now is being called "Job Auditions" by Ron Friedman, which is basically a contract with option to hire. I think it is a good additional approach for companies to get the best possible team and good for highly qualified candidates who are not so great at job in-person interviews.
A more specific example: close to a thousand job openings at Apple Inc. with specific search as "software engineer" and location as "Santa Clara Valley" -> Showing 921-939 of 600+
"During 2016, the VC sector provided $24.9 billion in financing, down 27.6 percent from the $34.4 billion in venture investing in 2015, according to the MoneyTree Report by PricewaterhouseCoopers and CB Insights."
I simply have to believe that cloud services have taken a big bite of the sysadmin market. We currently employ zero admins (~25 employees, moderate traffic).
In my previous ventures that'd be unthinkable, but we've effecticely outsourced the position to Amazon.
That just means you've elected not to hire an operations person, not that Amazon is doing it for you. You're still doing it, you just don't realize it.
This is very true: Dev/Op's has turned into an excuse to throw another job onto programers, one that in some circumstances they aren't really qualified to do.
We spent years as an industry separating out roles and building walled gardens. The latter was bad but the former was correctly informed.
You're presumably still handling package installation, deployment etc. via something handled in-house, no?
Certainly the "tending hardware" part of sysadminry is outsourceable, but unless you're going full-serverless there's still a bunch of systems work to do - so it may be longer before a full time sysadmin is worth the money, but one will likely be value for money eventually.
Yes but the ops teams at aws are not manned by magic unicorns. Also do be honest I have noticed that on teams where most ops are handled by devs if you convert that time to money it would have being at least break even to have an ops person even if part time with a much better results.
We're talking about 4,400 jobs here - how much of that can be attributed to one or two of he major layoffs announced in the past few quarters? If Intel or Cisco dropped a couple thousand of their layoff employees in San Jose, thats a pretty big chunk of the dip right there.
>"Global Equities Research’s Chowdhry believes the squeeze on the region’s tech sector is only beginning.
“Silicon Valley is bracing itself for literally a repeat of 2001,” Chowdhry said, referring to the dot-com meltdown that erupted with mammoth job losses that year. “This time it’s a social media bubble that will blow up."
The sky is falling! People are literally bracing themselves with both hands as we speak!
Nothing like using hyperbole to get you and your company's name in print. Global Equities Research? Seriously who the hell is that?
How many of these 4,400 jobs that the local economy has shed in SV are HP and Cisco which are two old guard tech companies not startups. These are also two companies who have been struggling with their old business models.
- IT jobs as in do they mean IT admins/Biz Analysts or programmers and software engineers. There is a smaller market in the area for IT admins than for programmers and software engineers. additionally, It seems to mix data between tech jobs and hotels etc.
- Why the random cheap shot at H1-Bs. Is this focused on employment or immigration ? typically in a healthy market employment trends follow immigration (employment based) trends. Seems like an effort to stir up some traffic to a poorly sourced article.
That was not a shot taken by the article, but by people who couldn't find jobs. H-1Bs are an easy scapegoat - certainly easier to explain your lack of success in finding a job because of H-1B visa holders rather than your own lack of in-demand skills.
I think you should consider that they may well be right, and that they can't find work at a rate they expect due to H1Bs working for less. People tend to know more about their own lives than you do about theirs.
You're missing/not addressing two important points.
First, as I pointed out, it's much easier cognitively to blame an external circumstance than your own skills deficit.
Second, any sort of expansion of the labor pool will tend to depress wages. If you can't find work at a rate you expect, you need to either lower your expectations or expect unemployment.
> If you can't find work at a rate you expect, you need to either lower your expectations or expect unemployment.
While this is true, if the rationale behind H1B program is designed to supplant a shortage in actual labor skills, and not in labor skills at the company owner's desired pay rates, and there is a surplus of skills due to unemployment while H1B's continue for the same skillset, then H1B's are compounding the problem due to misuse. This is logic.
Still waiting for the day that owners are clamoring for H1B's
for executives due to excessive pay and 'labor shortage'.. except executives are largely part of the same group as owners and the same people that fund the lobbyists so we all know that will never happen..
> if the rationale behind H1B program is designed to supplant a shortage in actual labor skills,
The intent of the law was to supplant shortages in labor skills, and that's the way the law is being used by most large "reputable" companies, not body shops.
I think you'd find it difficult to argue that when one of the big 5 brings somebody over on a H1b, it doesn't tend to depress the wages of engineers, as they usually only hire and recruit a highly selective (not necessarily talented) subset of engineers and compensate them extremely well.
The way the law is written allows a lot of consulting companies to game the system by bringing marginally talented labor to drive down the wages for skills that already exist in this country.
As an aside, this sort of argument against some H1-B seekers is a form of rent-seeking. You're trying to extract value from something that you inherited, e.g. your citizenship and your ability to legally work in this country, vs something you're able to do uniquely well.
Doubt it. I've seen a couple instances of experienced devs spooling up some interesting and objectively valuable machine learning systems, going from a lunchroom chat to production code in a very short amount of time. If there is a winter, it will be the result of AI becoming an off-the-shelf commodity; on the other hand, it should be easier to justify investments in research, given the demonstrated value of ML today.
The AI winter of the 80s was caused by over-hype, hardware companies whose markets were too small and whose products quickly fell behind the Moore's Law curve, and funding cuts by the US government.
We're mining a rich vein of techniques that have been opened up by the deep learning revolution. There doesn't seem to be any near-term end in sight, and the pace of improvement is incredible. It's hard just to keep up with the best result in, say, computer vision or NLP.
This is nothing like the 80's where techniques foundered on non-toy problems.
We're in Mountain View and we're hiring [0]. I've certainly noticed commercial real estate going from completely nuts to merely really expensive. Anecdotal sure. But eventually all the anecdotal points add up to a trend.
They specifically cite job losses in "IT" as a worrying indicator of the health of the tech industry. However, looking at the people quoted on LinkedIn, the problems seem to be for things like support/corporate IT - the people interviewed have worked at hospital helpdesks or as MS Access admins.
More indicative of general poor economic health and possibly higher ease of use or automation in these fields than of tech-industry-specific problems. Note that general poor economic health could itself be a leading indicator of trouble for our particular set.
I wonder what impact the Bay Area's cost of living and out of control housing prices had on this. I believe the tech scenes in other cities like LA, Seattle, NYC, Austin, etc have higher growth rates than the Bay Area.
Huge impact. Housing costs put a floor on acceptable salaries. Even if you go way out to San Jose or beyond, the housing costs and resulting monthly mortgage makes pushes minimum salaries into mid-150s for the most meager of family accommodations. I haven't moved from NYC to SF/SV for that singular reason.
Everybody uses words in their own way, but i'm pretty sure stumble is to almost fall. Like a momentary loss of balance. Tumble is worse, it's actually eating it.
Metaphors suck though. I don't know if 'balance' has any real meaning in the context of a job market. Supply and demand can't ever really be out of balance, they can just be distorted by random effects. In this sense 'fall' must mean smaller market?
Headlines are by nature vague and imprecise, they're just enough of a hook to make you decide to read the article or not. shrug
113 comments
[ 2.0 ms ] story [ 236 ms ] threadhttp://www.labormarketinfo.ca.gov/file/lfmonth/sjos$pds.pdf
“Unemployment rates throughout the Bay Area are among the lowest in the state. Almost anywhere else in the country, the Bay Area’s job trends would be the envy of that region.”
Every downturn has to start somewhere, and maybe this is the beginning of a bad stretch, but the largest employers are doing fine based on stock prices. Apple, Alphabet, Cisco, Facebook all made new all-time highs within the last month (or week).
Here is the March 2017 unemployment report. 3.5% for Santa Clara County. And states that jobs are + 2,000 from January to February.
http://www.labormarketinfo.edd.ca.gov/file/lfmonth/sjos$pds....
Profits -> hiring decisions -> stock prices
Stock price could be many months behind the profits as you don't do dialy earning reports.
Product announcements -> sales -> stock price -> reported profits -> hiring decisions (-> regional migration -> housing prices).
A number of my college classmates went into equity analysis or management consulting positions, they would do things like canvas retail locations and count the number of shoppers walking out with the products of interest, or track search engine rankings, or sit on Google Trends to watch for changes in search volumes in the markets of interest.
This is also why stock prices fall when a company "misses estimates", even if earnings are themselves up. Wall Street has a consensus estimate for how they expect the stock to do, and that estimate is priced into the stock. When the analysts are wrong (which isn't uncommon), the stock corrects based on the difference between the estimate and the reported earnings.
Stock price attempts to predict earnings. It is never perfect or earning announcements would never move the stock.
The fact is, if you want to know at a 1 or 2 month with window how a company is doing, stock is not the place to look
The problem is, can you exactly time when it is supposed to bust? Not likely. Markets can stay irrational longer than you can stay solvent.
This.
The numbers don't really back the first part of the statement up...
There are +3000 more I.T. jobs from 2016->2017 and I.T. only lost -500 jobs since last month, but that is a single month while the numbers are still higher than last year (down 0.7% monthly but up 4.1% overall yearly).
And "Computer Systems Design & Related Services" also grew 3.7% year over year with 0% loss in the last month.
The news is always looking for the negative 'sky is falling' angles and they will always be able to find someone to provide a sound bites to back up the story. But as usual the numbers say otherwise or are simply neutral/boring.
Some people are obsessed with being the first one to call the new demise of tech, or the latest bubble. But if anything the industry has become more resilient. Which is what happens when industries mature and grow larger. With more diversity and a larger base of employers/employees then fluctuations are naturally going to be smaller and better contained. That plus greater domain expertise develops over time - companies/investors learn how to manage risk better and understand the nature of the marketplace so overall losses are smaller.
https://www.dol.gov/whd/regs/compliance/FactSheet62/whdfs62M...
I think training is part of the problem. More importantly, workers need consulting help around how to take existing skills and use them to apply to upmarket jobs with higher pay.
Nah, it's an IQ issue. It's really not a (short-term) training issue, at least.
/i-will-burn-in-hell
I think a great piece of advice for life is, "Never assume you are the smartest person in the room." Remember that everyone brings something to the table.
DevOps/infrastructure is knowing how to glue together off the shelf tools and understanding how everything works. Great that as a dev you can probably bolt these tools together with docs, but your lack of context with the underlying fundamentals will eventually bite you.
And, thankfully, no architects.
Your comments convey a very narrow perspective of the job market that falls under "DevOps". That is to say, the sphere of roles looking to be filled by companies who believe they are filling a "DevOps" role is much larger than what you believe. It almost seems as if you are limiting yourself and projecting that onto the entire job market.
Now, I'm not sure you consider your current position as "limiting".. However, I can say that there are MANY roles being filled for "DevOps" candidates that extend beyond writing Ansible roles or Chef cookbooks. Speaking as someone 6 years into their engineering career and as the person who wrote the initial powershell(as a linux cloud engineer!) feature provider for Chef, I would not have taken a role such as you describe after year 1.
Now, you may be in a position where writing Ansible or Chef resources long term sounds swell. That's fare! But there are many, MANY roles that fall into the systems engineer/SRE bucket that are labeled at DevOps! Honestly, they are a matter of degree! The higher-end roles are senior, lead, unique, and require heavy development skills. If it's not the provisioning platform you are developing and supporting, it's the performance characteristics of the product itself.
That's actually not that much removed from quite a few programming jobs. It's mostly different manuals and the occasional plug to fit into a socket but other than that the differences are smaller than the similarities and debugging skills are important for both of them.
Unless you are arguing that people never did large scale automation of processes around deploying systems and software until this term existed...
In my personal vocabulary - there were always 2 kinds of sysadmins:
- Real sysadmins- which is now the term 'DevOps'
- IT System Technicians - which is what sysadmin is being perverted into
The old-style breakdown typically depended on how systems/development minded the company was - with more forward looking companies seeing their sysadmins as the former category and people who "administered" e.g. 'tended to' the IT "system", and the reactive companies who viewed IT as some foreign expense to be managed viewing their sysadmins as advanced IT techs who were 'admins' (in the administrative assistant sense) for the various 'systems' (e.g. computers) that they purchased.
As someone who always operated in the mindset of the former category, but unfortunately has operated mostly in the latter type of company, it is difficult to make the transition into 'DevOps' roles since this term has been subsequently invented to describe the work that I always did, using tools that I would have written from scratch myself in the old days, or deployed myself in the current days, but unfortunately due to crappy management and backwards modes of thinking, don't have the proper 'current' job-title on my resume (e.g. sysadmin)
If you talk to someone about their day and it revolves around tickets, they are the techs that you're talking about. The real "sysadmins"/devops understand systems and own things.
It's ok to realign your title with function. Every company handles this sort of thing differently. I was a "Computer Systems Programmer 3" for two years. In that organization, that rank was roughly equivalent to a Staff Engineer. My resume says "Lead DBA", as that was my primary function... the formal title assigned was just whatever was available when I was promoted... the title description was written for mainframe people!
> DevOps/infrastructure engineer is basically a programming position
sysadmin/network admin ==>> DevOps
IT manager => Manager or lead.
DevOps is not a programming position at all. It is usually cheap system administration (that replace the term sysadmin) and it does NOT require coding. Pay attention to the name and the place you joins, you're can easily be burnt by joining a team of operations monkeys while thinking they were the dev and system type persons.
Personal services IT is still alive and well and a number of people I know have made a living out of helping older people manage passwords, upgrade their systems, move their phone plans or transfer their data when they get a new phone etc. But to be successful at that you also need to talk to people and be able to maintain a business relationship with them, not a skill that everyone has in addition to their deep knowledge of IT.
As tm2d mentions devops is still a hot job market. But it is not the small business 'tech' role so there are fewer actual slots for that role.
There was also a comment in the article about H1-B visas and employers wanting "younger and less expensive" workers. I try to remind my older friends that if someone can spend 6 months learning to do what you do and do it well enough to meet the needs of the job, then you are only "worth" what a company would pay that person they just hired. If you want to have a larger salary and better job security, then you need to be able to do things that can't be trained in 6 months. The days of 'too few programmers to go around' are long past, there is now a surplus and they are coming fast and furious out of college.
Some of your other points were insightful, but this assertion contradicts my experience and intuition, and -- I believe -- labor market statistics.
In 1999 when I was hiring programmers and offering above market starting salaries I wouldn't get any qualified resumes.
It is from that experience that would assert it is a 'pricing' issue rather than a 'selection' issue.
[1] After 10 qualified I had my first interview round, 3 were brought back for secondary interviews and one hired. I expect there were easily 20 - 30 engineers in there that could have done the job.
It seems that there might be plenty of churn overall.
We've gone through surplus years before, and they don't look like this.
But to management/high ups, they "can do the job" and are cheaper. It's like outsourcing. You save in costs but there are hidden downsides and risks.
I'd probably argue that very few engineer jobs can be done competently with a few months of training.
Coding is a cheap easy skill that 15-20 years old can do.
Executing and managing a project from start to finish is a rare skill that cannot be acquired in school or the internet.
Coding Monkey => Cheap and easy to find.
Coder who has successfully led and executed projects => Expensive and rare.
I eventually figured out there were very limited opportunities for someone with my skills and interests in that role so I studied and worked to build the skills to get a software engineering job.
I think that Helpdesk roles in general should be seen as stepping stones toward careers in more advanced roles. I actually think my experience doing that work has made me a better engineer and coworker.
To backup my "significant number" claim, here's a 3 months old info: "Bay Area unemployment rates fall to lowest in 15 years" http://www.mercurynews.com/2016/12/16/south-bay-jobs-surge-e...
A more specific example: close to a thousand job openings at Apple Inc. with specific search as "software engineer" and location as "Santa Clara Valley" -> Showing 921-939 of 600+
Can safely ignore bubbles created by VC funding cash-flow injection and focus on business that bootstrap in a healthy way.
$10B and 28% are hard to ignore.
[1]:https://www.nytimes.com/2016/06/02/technology/uber-investmen...
In my previous ventures that'd be unthinkable, but we've effecticely outsourced the position to Amazon.
We spent years as an industry separating out roles and building walled gardens. The latter was bad but the former was correctly informed.
Certainly the "tending hardware" part of sysadminry is outsourceable, but unless you're going full-serverless there's still a bunch of systems work to do - so it may be longer before a full time sysadmin is worth the money, but one will likely be value for money eventually.
With immutable infrastructure all defined in config file (cloudformation, terraform, etc), you don't have any pet servers sitting around.
“Silicon Valley is bracing itself for literally a repeat of 2001,” Chowdhry said, referring to the dot-com meltdown that erupted with mammoth job losses that year. “This time it’s a social media bubble that will blow up."
The sky is falling! People are literally bracing themselves with both hands as we speak!
Nothing like using hyperbole to get you and your company's name in print. Global Equities Research? Seriously who the hell is that?
How many of these 4,400 jobs that the local economy has shed in SV are HP and Cisco which are two old guard tech companies not startups. These are also two companies who have been struggling with their old business models.
- IT jobs as in do they mean IT admins/Biz Analysts or programmers and software engineers. There is a smaller market in the area for IT admins than for programmers and software engineers. additionally, It seems to mix data between tech jobs and hotels etc.
- Why the random cheap shot at H1-Bs. Is this focused on employment or immigration ? typically in a healthy market employment trends follow immigration (employment based) trends. Seems like an effort to stir up some traffic to a poorly sourced article.
- The social media bubble? This is the first time I have heard of that term. Appreciate if they did some basic homework to see 2016 investment trends https://www.wilmerhale.com/uploadedFiles/Shared_Content/Edit...
That was not a shot taken by the article, but by people who couldn't find jobs. H-1Bs are an easy scapegoat - certainly easier to explain your lack of success in finding a job because of H-1B visa holders rather than your own lack of in-demand skills.
First, as I pointed out, it's much easier cognitively to blame an external circumstance than your own skills deficit.
Second, any sort of expansion of the labor pool will tend to depress wages. If you can't find work at a rate you expect, you need to either lower your expectations or expect unemployment.
While this is true, if the rationale behind H1B program is designed to supplant a shortage in actual labor skills, and not in labor skills at the company owner's desired pay rates, and there is a surplus of skills due to unemployment while H1B's continue for the same skillset, then H1B's are compounding the problem due to misuse. This is logic.
Still waiting for the day that owners are clamoring for H1B's for executives due to excessive pay and 'labor shortage'.. except executives are largely part of the same group as owners and the same people that fund the lobbyists so we all know that will never happen..
The intent of the law was to supplant shortages in labor skills, and that's the way the law is being used by most large "reputable" companies, not body shops.
I think you'd find it difficult to argue that when one of the big 5 brings somebody over on a H1b, it doesn't tend to depress the wages of engineers, as they usually only hire and recruit a highly selective (not necessarily talented) subset of engineers and compensate them extremely well.
The way the law is written allows a lot of consulting companies to game the system by bringing marginally talented labor to drive down the wages for skills that already exist in this country.
As an aside, this sort of argument against some H1-B seekers is a form of rent-seeking. You're trying to extract value from something that you inherited, e.g. your citizenship and your ability to legally work in this country, vs something you're able to do uniquely well.
That's a new type of H1-B visa fraud that I had never heard of as far as I can tell from the article.
https://www.linkedin.com/pulse/ai-winter-coming-sandro-skans...
so I don't expect deep learning to save us.
No one knows who that is.
Namedropping in an authoritative tone about a random person, seems kind of odd.
The AI winter of the 80s was caused by over-hype, hardware companies whose markets were too small and whose products quickly fell behind the Moore's Law curve, and funding cuts by the US government.
General-purpose chatbot, almost equal to solve AGI. If that is how you define winter...where is summer!?
This is nothing like the 80's where techniques foundered on non-toy problems.
https://jobsquery.it/jobs;page=1;tags=;sortBy=PUBLISHED_AT_D...
[0]We're 18 people, enterprise SaaS, started by 2 founders who have worked together for 15 years. https://www.simplelegal.com/careers
Also looking for Head of Engineering and devs (senior front end dev, lead API dev, Django dev).
More indicative of general poor economic health and possibly higher ease of use or automation in these fields than of tech-industry-specific problems. Note that general poor economic health could itself be a leading indicator of trouble for our particular set.
Metaphors suck though. I don't know if 'balance' has any real meaning in the context of a job market. Supply and demand can't ever really be out of balance, they can just be distorted by random effects. In this sense 'fall' must mean smaller market?
Headlines are by nature vague and imprecise, they're just enough of a hook to make you decide to read the article or not. shrug