Posts Tagged ‘technology’

Semiconductor Subsidies: The Wrong Solution For The Wrong Problem

Thursday, January 20th, 2022

There’s no problem that the federal government throwing money at it can’t make worse.

Today’s example: Democrats pimping billions in taxpayer subsidies for the semiconductor industry.

As the COVID-19 pandemic exacerbates supply chain backlogs and global computer chip shortages

Correction: It wasn’t the pandemic itself, it was government lockdowns and other overeactions that did that.

Democratic leaders in Congress as well as President Joe Biden want Congress to fast track a $250 billion bill to develop American independence from China and other competitors in chip manufacturing.

The Capital Region – home to SUNY Polytechnic Institute, the only publicly owned 300-millimeter semiconductor research and development center in the U.S. – stands to reap significant benefits from the enactment of Senate Majority Leader Charles E. Schumer’s multi-billion dollar bill, which he envisions as a direct investment in his home state’s economy.

“Sen. Schumer wrote this legislation with upstate New York always at the forefront of his mind,” Schumer’s spokeswoman Allison Biasotti said. “We are already seeing the excitement in major employer expansions and thousands of jobs on the horizon from GlobalFoundries’ planned expansion (in Malta) and (his) push for Albany Nanotech to be a hub for the National Semiconductor Technology Center.”

A focal point of the bill, which the New York Democrat co-sponsored with Sen. Todd Young, R-Ind., is a historic $52 billion investment in stateside semiconductor research and development to address a global chip shortage plaguing the automotive industry.

Lawmakers began to focus more on the low domestic production of semiconductors when the COVID-19 pandemic cut off supplies from overseas. Without access to chips, several automakers shut down their production lines, and manufacturers of essential medical devices and consumer electronics struggled to meet increasing demand.

Roughly 12 percent of the world’s semiconductors are manufactured in the United States, down from 37 percent in 1990, according to the Semiconductor Industry Association.

Either these stats are false or misleading (probably the latter). The most recent stats I can find show that the United States has some 47% of the semiconductor market. It’s possible that the 12% refers to the entire worldwide number of individual chips produced, including discrete components (transistors, resistors, etc.). Those are indeed semiconductors, but they’re produced on old amortized fabs (inside the industry these are referred to as “jelly bean factories”) and sell for pennies a piece (or less). If you’re already in that industry, those old fabs make small, steady profits every year, but nobody jumps into that business with new fabs.

The chips China make are generally either: A.) Cheap, or B.) intended for their internal market. No one sends cutting edge chips to be fabbed in China because they don’t have the tech to do it and everyone know they’ll steal your designs and crank out knock-offs on the sly whenever possible. China’s semiconductor industry is mostly smoke and mirrors all the way down.

Semiconductor subsidies have all the hallmarks of a classic Washington boondoggle: The wrong action at the wrong time for the wrong problem.

First, there are already signs that the automotive semiconductor crunch is easing, thanks not to the Biden Administration but to the actions of the free market.

Second, the shortage wasn’t the result of a “chip shortage,” it was the result of “a lack of available foundry wafer starts.” Automakers cancelled their orders for display drivers when it looked like Flu Manchu lockdowns were going to depress the economy for a while, and were caught off-guard by the V-shaped recovery under Trump, and got sent to the back of the line to get their product fabbed after they changed their mind. Remember, just about all foundries are running flat-out 24/7/365, pausing only to switch to different chips for different customers. There’s no slack in the system, and those wafer starts are already spoken for (and possibly paid for) by other customers well in advance. Just as nine woman can’t give birth to a fully grown baby in one month, you can’t just “make chips quicker” in an existing fab.

Third, remember that cutting edge semiconductor fabs are hideously expensive. Moore’s second law states that the cost of a new, cutting edge semiconductor plant doubles every four years. Samsung’s planned fab in Taylor, Texas is going to cost $17 billion.

Fourth, if you go to a random semiconductor company and go “Here’s 20 billion! Go build a state-of-the-art 5nm wafer fabrication plant!”, then:

A.) You’re looking at a very minimum of 2-3 years before the first production wafer comes off the line. You can’t just take an existing building and turn it into a fab, it has to be specially built from the ground up with exacting standards for cleanroom air filtering, concrete slab level uniformity, etc. And 2-3 years is probably the lead time to get an ASML EUV stepper.

B.) Unless you’re TSMC, Samsung or (maybe) Intel, the answer is probably “Uh, we’ll try, but no promises,” because those three companies are the only ones that actually having wafer fabs running 10nm or smaller process nodes. GlobalFoundries, mentioned in the article, has Fab 8 in Malta, NY, running 14nm, which is not horribly far off the state-of-the art, but not good enough to fab the really cutting-edge chips demanded of companies like Apple, NVIDIA, etc. Tiny problem: In 2018, GlobalFoundries stopped all work on 7nm development.

The contract maker of semiconductors decided to cease development of bleeding edge manufacturing technologies and stop all work on its 7LP (7 nm) fabrication processes, which will not be used for any client. Instead, the company will focus on specialized process technologies for clients in emerging high-growth markets. These technologies will initially be based on the company’s 14LPP/12LP platform and will include RF, embedded memory, and low power features.

So it was too hard a game for them to play, but with a big heap of taxpayer subsidies, I’m sure they’d be willing to give it another go.

Of course, you don’t need a cutting edge fab to build display drivers. Bosch just opened a $1.2 billion, 65nm fab in Dresden to do just that. But you don’t need subsidies to build trailing edge fabs.

$250 billion in taxpayer subsidies wouldn’t get you a single additional wafer start this year, and probably would accomplish little more than channeling money to politically connected firms and sticky pockets in a state (New York) that no one wants to build fabs in any more because of high costs, high taxes and union rule requirements.

It’s a bad idea congress should reject.

Log4J and Internet Castles Made of Sand

Thursday, December 16th, 2021

If you work outside of a tech company, chances are you’ve spent this week primarily concerned with getting ready for Christmas. If you work inside a tech company, there’s a significant chance your company spent much of this week patching a critical vulnerability in an open source Java logging library called Log4J.

Here’s a non-technical explanation of the problem:

It’s a vulnerability that was discovered in a piece of free, open source software called log4j. This software is used by thousands of websites and applications, to perform mundane functions most people don’t think about, such as logging information for use by that website’s developers, for debugging and other purposes.

Every web application needs functionality like this, and as a result, the use of log4j is ubiquitous worldwide. Unfortunately, it turns out log4j has a previously undiscovered security vulnerability where data sent to it through that website — if it contains a special sequence of characters — results in log4j automatically fetching additional software from an external website and running it. If a cyberattacker exploits this, they can make the server that is running log4j run any software they want — including software that can completely take over that server. This is known as a Remote Code Execution (RCE) attack.

To use a technical phrase, this is Really Bad.

The net result is that, left unaddressed, cyberattackers right now can completely take over thousands of websites and online applications, allowing them to steal money, data, and access. The security community has been completely focused on this vulnerability for the past two days, and updating servers running log4j as quickly as possible to protect against this vulnerability.

The good news is that mitigations are relatively easy to implement. The bad news is that left unmitigated, the vulnerability is extremely easy to exploit. iCloud, Minecraft, Baidu, and many other sites have been confirmed to be vulnerable so far, and you’ll likely hear more about many other sites being vulnerable in the coming days.

And those companies are just the tip of the iceberg. LAMP stacks (Linux, Apache, MySQL, and PHP) are used as the technological underpinnings for a wide variety of web applications of all sizes. (It’s not universal, as NGINX has taken over as a market leader from Apache, and there are still a few all-Microsoft houses that use IIS, and neither of them have the vulnerability.)

Open Source has been a revolutionary invention because it provides rapid development by armies of distributed developers, and Linus’s Law states that “with enough eyes, all bugs are shallow.” But there are tens of thousands of Open Source components out there running critical infrastructure that haven’t had nearly as many eyes on the code as the Linux kernel. It’s simply the nature of the beast. XKCD had a cartoon for this occasion:

Internet applications gain usefulness from widespread adoption and the number of other components they tie into and support. You know what creates new vulnerabilities? A larger user base and the number of other components they tie into and support, which creates more attack surfaces for malicious actors to exploit.

The flaw isn’t the fault of Random Guy in Nebraska, the fault is the company adopting software that they can’t possibly test for all the use-cases they’re going to use it for. Surprise! Just about every high tech company in the world is in the same boat. Pretty much everyone uses a wide panoply of open source tools for their Internet applications, and no one can test all the permutations of how each component might be put to use.

You can’t eliminate the risk, you can only minimize and mitigate it. You can use containerization strategies (Docker, Kubernetes, Container D, etc.) to minimize attack surfaces and limit contagion. You can run all your code through security scanning tools on your CI/CD platform of choice. You can do constant testing and keep rolling backups of everything to limit risk and speed recovery. (You can also train your employees not to click on random email links without verifying the sender is who they say they are, and not to give any any account information or passwords over the phone, and train them enough so that the lessons stick, even though phising and human engineering weren’t factors in the Log4J vulnerability.)

But there still a good chance that the platform you’re using today is different than the platform you’ll be using ten years from now, and you’ll have to go through the same learning lessons discovering new vulnerabilities for the new platform all over again.

Castles made of sand all fall into the sea eventually…

Samsung To Build $17 Billion Fab in Taylor, Texas

Tuesday, November 23rd, 2021

Reports indicate that semiconductor giant Samsung has picked Taylor, Texas as the site for a $17 billion wafer fabrication plant.

In recent days, Williamson County and the city of Taylor had seemed to emerge as the frontrunner to land a $17 billion chipmaking plant planned by Samsung.

Now, it seems the technology giant has indeed picked the small Central Texas city as the site for its next major operation, according to media reports.

Citing unnamed sources with knowledge of the decision, the Wall Street Journal reported Monday night that Samsung has picked Taylor over sites in Austin, Arizona and New York.

Samsung has not formally confirmed the decision, and a company spokesperson did not immediately respond to messages left by the American-Statesman on Monday evening. However, the announcement is expected to be made in a news conference with Gov. Greg Abbott at the Texas Capitol on Tuesday afternoon.

If Samsung does, in fact, build the facility at the Taylor site, it will be the latest in a stunning run of economic development wins for the Austin area, and for its technology sector in particular.

Tesla announced Oct. 7 that the automaker will move its corporate headquarters from California to Austin. That news came 15 months after Tesla chose an Austin-area site as the home for its $1.1 billion manufacturing facility. Software giant Oracle announced last December that it was moving its corporate headquarters from California to Austin, and a number of other technology giants — including Apple, Facebook, Google and Amazon — have recently expanded their operations in Central Texas.

Samsung recently overtook Intel as the largest semiconductor manufacturer in the world, and along with TSMC, those three are also the only real players in cutting-edge under-10nm processes. As I’ve mentioned before, new cutting edge fabs are hideously expensive to build. TSMC is a foundry (which means they fab other people’s chip designs), while both Samsung and Intel are integrated device manufacturers (IDMs), meaning they fab their own designs, though I think both dabble in foundry work as a sideline. (Samsung is also one of the largest flat panel screen manufacturers in the world; flat panel manufacturing uses semiconductor manufacturing techniques, but is fundamentally a different industry, and just about all flat panels are produced in Asia these days.)

The decision to eliminate New York from the list was probably quite easy. Back when IBM was running it’s state-of-the-art fabs in East Fishkill, there was considerable technological infrastructure in the state. Back In The Day IBM had some of the most respected process technology knowledge in the industry. But then they got out of the manufacturing business, and the East Fishkill fab got sold to Global Foundries, who later sold it to ON Semiconductor. But today New York constantly ranks among the worst states in the nation for business environment, due to high taxes, excessive regulation, and the gradual decay of infrastructure and institutions that comes with one-party Democrat control.

Arizona is a much stronger candidate. Intel has a huge complex of modern fabs in Chandler and TSMC is building a state of the art fab in Phoenix proper, which means there’s a lot of local talent and infrastructure to draw on. A purple state, Arizona usually ranks in the top ten for a business-friendly climate, but they do have a personal income tax.

Texas, by contrast, is constantly rated as the top or second best business climate the the country (occasionally losing to Florida), and has no state income tax. Samsung already has a fab in Austin, along with older legacy fabs from NxP (ex-Motorola) and Infineon, along with significant presence by the major semiconductor equipment manufacturing giants (Applied Materials, Tokyo Electron, etc.). Taylor is close enough to Austin to draw on the technical talent and infrastructure there, without having to worry about the crazy left-wing politics, as Williamson County, while having turned a bit more purple lately, is still safe Republican territory.

Another solid reason to locate in Taylor: ERCOT is headquartered there, which means the area will never be power-cycled in an emergency. The winter storm evidently cost Samsung $268 million in lost revenue from the outage, which I can well believe. When the power goes off, all the equipment needs to be requaled, which is a long, painful process for a single machine, much less the some 200+ needed in a modern fab.

America has lots of tech hubs: Silicon Valley, Seattle, the North Carolina triangle, greater Boston, etc. But nobody is building cutting edge fabs in those areas. Central Texas has rapidly expanding software, hardware and silicon industries.

Austin is primed to be one of the greatest global tech hubs of the 21st century, assuming Austin political leadership doesn’t screw it up…

Drone Swarm Boogaloo

Sunday, November 14th, 2021

Let’s talk about drone swarms.

Relatively cheap, quickly deployed swarms of autonomous drones are probably going to be a major factor in short- to -medium-term warfare. There will probably be (at least) two different types of autonomous drones: Suicide drones for hard targets like tanks, and anti-personnel drones using light weapons. The later could either return to base, or just fall to the ground for later retrieval and refurbishment when their fuel or power run out.

Both will pick out targets using AI.

I’m not much interested in the central question posed by the following video (are drone swarms technically WMDs), mainly because China doesn’t give a rat’s ass about international law. But it shows a variety of different drones being developed in various countries:

Speaking of China, here’s a short video on China deploying drones via a MRLS:

China is investing not only in drones, but also in counter drone technology.

Here’s a look at the Navy’s LOCUST system from four years ago:

LOCUST had a successful test earlier this year.

The advantages of functional drone swarms for armored or naval warfare should be obvious. If you can kill a $10 million Abrams or Type 99 tank and crews with a $100,000 drone, that’s a clear win. Whether such drones can overcome current active protection systems like Trophy is an open question. And Germany’s Rheinmetall just released a video of an anti-drone platform shooting some down:

The problem, of course, is that their system hasn’t demonstrated any autonomous mode, and real battlefield drones will probably quickly adopt a variety of evasive maneuvers rather than hovering nicely in a row to be shot.

Welcome to the AI drone arms race. Make your own SkyNet jokes in the comments below.

Joe Rogan and Ben Shapiro on The Virtual Metaverse

Thursday, November 11th, 2021

Joe Rogan and Ben Shapiro discuss Facebook’s announcement of a virtual reality “metaverse.” Some interesting discussion, though little that will be of surprise to anyone who read cyberpunk in the 1980s:

Shapiro: “I wonder with this stuff if we’re innovating ourselves out of existence as a civilization.”

Shapiro:”Let’s just take this on the most baseline demographic level. None of them get married, none of them have babies. In two generations, this ain’t gonna matter. You’re gonna have a good time in the virtual reality and then they’re gonna be no babies to carry this on and the only on earth are going to be religious Jews, religious Catholics and religious Muslims and that’s it.”

Knowing Facebook, I can only assume their Metaverse will be plastered with annoying ads and “features” you hate. Also, just as soon as you get comfortable in the Metaverse, they’ll change the interface…

Did Google Break The Law?

Sunday, October 24th, 2021

I know that headline is more than a little ambiguous, as Google has probably broken multiple laws, if only because they’re so big and there are so many laws. But “Did Google break the law using sneaky, underhanded means to carry out anti-competitive trade practices to kill off an alternative ad allocating system called ‘header bidding’ because it threatened to damage one of its biggest revenue streams” is way too long for a blog post title.

As a prelude, here’s a brief description of header bidding and how it differs from Google’s “Waterfall” system:

Header bidding is an advanced programmatic advertising technique that serves as an alternative to the Google “waterfall” method. Header bidding is also sometimes referred to as advance bidding or pre-bidding, and offers publishers a way to simultaneously offer ad space out to numerous SSPs or Ad Exchanges at once.

Normally, when a publisher is trying to sell advertising space on its site, the process for filling inventory goes something like this:

First, your site reaches out to your ad server. In general, direct-sold inventory takes precedence over any programmatically sold options. Next, available inventory is served through the site’s ad server, such as Google DoubleClick in a waterfall sequence, meaning unsold inventory is offered first to the top-ranked ad exchange, and then whatever is still unsold is passed along to the second ad exchange, and so on. These rankings are usually determined by size, but the biggest ones aren’t necessarily the ones willing to pay the highest price. (For publishers, this means lower overall revenue if the inventory isn’t automatically going to the highest bidder.)

To further complicate the process, sites using Google’s DFP for Publishers has a setting that enables them to outbid the highest bidder by a penny using Google Ad Exchange (AdX). And since AdX gets the last bid, they are generally in a position to win most of these auctions.

Publishers end up feeling like they aren’t making quite as much money as they would without Google meddling in the bids.

How Does Header Bidding Help Publishers?

Header bidding is a way for publishers to have a simultaneous auction from all the bidders, rather than the sequential strategy that Google uses. By placing some javascript on their website, when a particular page is loaded, it reaches out to all supported SSPs or ad exchanges for bids before its ad server’s own direct-sold inventory is called. Publishers can even choose to allow the winning bid to compete with pricing from the direct sales.

Got that? Here, as best I can understand, is a summary example:

Say Joe Blow’s Ad Agency and Attack Lawyer Collective wants to be the top bidder for serving ads up for the keyword “mesothelioma” (which, at one time, was the priciest keyword you could buy for digital ads), and it is willing to pay, say, $100 per 1,000 impressions. Under Google’s waterfall method, they would never get to bid if Big Madison Avenue Ad Agency was in the top tier of bidders even though BMAAA only offered $50 per 1,000 impressions, because Google would sell those ad slots only to the highest bidder in the top tier, and would never get down to Joe Blow in the third tier. (This is all greatly oversimplified, and feel free to correct/amend this example in the comments.)

Well, due to the big antitrust lawsuit filed against Google by some 38 (last time I looked) state attorney generals (including Texas), lots of dirty secrets and memos have come to light as part of discovery. Many of the most serious bits were redacted, but that was just changed by judge’s orders:

Two corporate behemoths getting together to strike insider deals with each other that freeze out competitors is pretty much textbook anti-competitive practices 101 stuff.

Holy shit! Google and Facebook are agreeing not to cooperate with any antitrust action by the federal government to bring action against the other. That’s not a red flag, that’s the Nostromo‘s flashing lights and screaming self-destruct klaxon in the original Alien.

So according to these documents, Google is not only a monopoly, it is a coercive monopoly that uses illegal anti-competitive trade practices to stifle competition.

And since the lawsuit was brought by a bipartisan coalition of state attorney generals, Google can’t just buy a few tens of millions of dollars worth of Hunter Biden painting to make the entire thing go away…

The System is Down…The System Is Down…

Monday, October 4th, 2021

Facebook, Instagram and WhatsApp are all down right now. Botched upgrade? Misconfigured router? Expired signing certificate? Who knows? I’m just going to assume its a problem with their latest SuppressAllPostsQuestioningTheHolyVaccineMarrative.yml file. But it’s a reminder of how deeply interconnected all online systems are these days, and how many different things can go wrong at different layers.

Expect a sudden burst of productivity from American companies.

And just in case you didn’t get the reference:

Edited to add: Additional detail:

Facebook—and apparently all the major services Facebook owns—are down today. We first noticed the problem at about 11:30 am Eastern time, when some Facebook links stopped working. Investigating a bit further showed major DNS failures at Facebook…

DNS—short for Domain Name System—is the service which translates human-readable hostnames (like arstechnica.com) to raw, numeric IP addresses (like 18.221.249.245). Without working DNS, your computer doesn’t know how to get to the servers that host the website you’re looking for.

The problem goes deeper than Facebook’s obvious DNS failures, though. Facebook-owned Instagram was also down, and its DNS services—which are hosted on Amazon rather than being internal to Facebook’s own network—were functional. Instagram and WhatsApp were reachable but showed HTTP 503 (no server is available for the request) failures instead, an indication that while DNS worked and the services’ load balancers were reachable, the application servers that should be feeding the load balancers were not.

A bit later, Cloudflare VP Dane Knecht reported that all BGP routes for Facebook had been pulled. (BGP—short for Border Gateway Protocol—is the system by which one network figures out the best route to a different network.)

With no BGP routes into Facebook’s network, Facebook’s own DNS servers would be unreachable—as would the missing application servers for Facebook-owned Instagram, WhatsApp, and Oculus VR.

DNS—short for Domain Name System—is the service which translates human-readable hostnames (like arstechnica.com) to raw, numeric IP addresses (like 18.221.249.245). Without working DNS, your computer doesn’t know how to get to the servers that host the website you’re looking for.

The problem goes deeper than Facebook’s obvious DNS failures, though. Facebook-owned Instagram was also down, and its DNS services—which are hosted on Amazon rather than being internal to Facebook’s own network—were functional. Instagram and WhatsApp were reachable but showed HTTP 503 (no server is available for the request) failures instead, an indication that while DNS worked and the services’ load balancers were reachable, the application servers that should be feeding the load balancers were not.

A bit later, Cloudflare VP Dane Knecht reported that all BGP routes for Facebook had been pulled. (BGP—short for Border Gateway Protocol—is the system by which one network figures out the best route to a different network.)

With no BGP routes into Facebook’s network, Facebook’s own DNS servers would be unreachable—as would the missing application servers for Facebook-owned Instagram, WhatsApp, and Oculus VR.

Speculation is that Facebook engineers have locked themselves out of their own network, meaning someone with physical access to the servers will have to fix things…

Edited to add 2: Krebs offers more details:

Facebook and its sister properties Instagram and WhatsApp are suffering from ongoing, global outages. We don’t yet know why this happened, but the how is clear: Earlier this morning, something inside Facebook caused the company to revoke key digital records that tell computers and other Internet-enabled devices how to find these destinations online.

Doug Madory is director of internet analysis at Kentik, a San Francisco-based network monitoring company. Madory said at approximately 11:39 a.m. ET today (15:39 UTC), someone at Facebook caused an update to be made to the company’s Border Gateway Protocol (BGP) records. BGP is a mechanism by which Internet service providers of the world share information about which providers are responsible for routing Internet traffic to which specific groups of Internet addresses.

In simpler terms, sometime this morning Facebook took away the map telling the world’s computers how to find its various online properties. As a result, when one types Facebook.com into a web browser, the browser has no idea where to find Facebook.com, and so returns an error page.

In addition to stranding billions of users, the Facebook outage also has stranded its employees from communicating with one another using their internal Facebook tools. That’s because Facebook’s email and tools are all managed in house and via the same domains that are now stranded.

“Not only are Facebook’s services and apps down for the public, its internal tools and communications platforms, including Workplace, are out as well,” New York Times tech reporter Ryan Mac tweeted. “No one can do any work. Several people I’ve talked to said this is the equivalent of a ‘snow day’ at the company.”

Developing…

Edited to add 3: Seeing reports that Gmail is down for some people. It’s not down for me. I just tested and it’s working fine.

Updated to add 4: Facebook appears to be back up, but is way wonky…

A Big Chunk Of The Internet Went Down This Morning

Tuesday, June 8th, 2021

While you were asleep this morning, a big chunk of the Internet briefly went down:

Among those affected are Amazon, Twitch, Reddit, The Verge, The Guardian, ZDnet, The New York Times, Freetrade, The Financial Times, Pinterest, Kickstarter, Ebay, The Telegraph, CNN, and Imgur. Google searches are also partially impacted, as is the Google Cloud Platform. While Twitter is up, its emoji platform is offline.

The issue has been traced back to content delivery network Fastly, which is down. The company runs an Edge cloud between companies’ data centers and the end user, reducing latency, protecting from DDoS attacks, and helping them handle traffic spikes.

Fastly is a content delivery network (CDN), an intermediary that brings data closer to Internet end users so their interactions don’t need to go all the way back to the company’s central servers. This is the market Akamai pioneered, and other companies in the space include CDNetwork and Cloudflare. “Edge computing” is a sort of catch-all term for intermediary cloud services that became one of those buzzwords that VC companies threw money at about two years ago.

With Amazon down, a lot of people jumped to the conclusion that AWS, Amazon’s 800 pound cloud service gorilla, was experience an outage, but it turned out to be Fastly, who evidently fixed the problem at 10:57 UTC (5:57 AM CDT):

“The issue has been identified and a fix has been applied. Customers may experience increased origin load as global services return.” On Twitter, the company added: “We identified a service configuration that triggered disruptions across our POPs globally and have disabled that configuration. Our global network is coming back online.”

The modern Internet is decentralized, widely distributed and pretty efficient, but its very decentralized nature means that there are more moving parts to break, and also more attack surfaces for hackers to exploit. Delivering rich content over the Internet (be it text, images, video or shopping) usually involves dozens, if not hundred of software pieces, protocols, companies, etc. for every web page served up. Any of them can go down. Network engineers design in as much redundancy as possible, but there’s only so much you can do. I worked for a company in 2020 whose computer testing lab went down because antifa rioters in Minneapolis physically destroyed a fiber optic cable.

All I can tell you is to keep multiple rotating backups of your most valuable data, because anything that can go wrong eventually will…

They’re Not Going Back

Wednesday, June 2nd, 2021

There’s a very early Laurie Anderson song called “Walk The Dog” where she riffs on (among other things) a Dolly Parton song:

“I just want to go back to my Tennessee mountain home now.”

Well, you know she’s not gonna go back home.

And I know she’s not gonna go back home.

And she knows she’s never gonna go back there.

And that’s a good summary of many former office workers post-coronavirus: They’re never going back.

With the coronavirus pandemic receding for every vaccine that reaches an arm, the push by some employers to get people back into offices is clashing with workers who’ve embraced remote work as the new normal.

While companies from Google to Ford Motor Co. and Citigroup Inc. have promised greater flexibility, many chief executives have publicly extolled the importance of being in offices. Some have lamented the perils of remote work, saying it diminishes collaboration and company culture. JPMorgan Chase & Co.’s Jamie Dimon said at a recent conference that it doesn’t work “for those who want to hustle.”

But legions of employees aren’t so sure. If anything, the past year has proved that lots of work can be done from anywhere, sans lengthy commutes on crowded trains or highways. Some people have moved. Others have lingering worries about the virus and vaccine-hesitant colleagues.

And for [Portia] Twidt, there’s also the notion that some bosses, particularly those of a generation less familiar to remote work, are eager to regain tight control of their minions.

“They feel like we’re not working if they can’t see us,” she said. “It’s a boomer power-play.”

It’s still early to say how the post-pandemic work environment will look. Only about 28% of U.S. office workers are back at their buildings, according to an index of 10 metro areas compiled by security company Kastle Systems. Many employers are still being lenient with policies as the virus lingers, vaccinations continue to roll out and childcare situations remain erratic.

But as office returns accelerate, some employees may want different options. A May survey of 1,000 U.S. adults showed that 39% would consider quitting if their employers weren’t flexible about remote work. The generational difference is clear: Among millennials and Gen Z, that figure was 49%, according to the poll by Morning Consult on behalf of Bloomberg News.

“High-five to them,” said Sara Sutton, the CEO of FlexJobs, a job-service platform focused on flexible employment. “Remote work and hybrid are here to stay.”

The lack of commutes and cost savings are the top benefits of remote work, according to a FlexJobs survey of 2,100 people released in April. More than a third of the respondents said they save at least $5,000 per year by working remotely.

This is especially true in high tech. If you have in-demand skills (full-stack developer, AI expertise, etc.), lots of companies are vying for you, and all of them have remote-work infrastructure already in place. Chances are good you login into a VPN in the morning, communicate via email and Slack, have your meetings on Zoom, code on your laptop, then check your work into a remote repository running a continuous integration/continuous deployment platform (GitHub, GitLab, etc.) that builds and tests your software. There’s zero reason for you to spend your time commuting to the office. And if your current employer won’t let you work from home, another will. And that other company can be located anywhere, and they can hire the best talent for their position no matter whether they have a local office.

I, for one, save just shy of an hour a day working from home rather than braving Austin roads, and my dogs are much happier.

How can you keep them in the big city once they’ve tasted life back on the farm?

Assuming the farm has Internet…

A Good Explanation of the Semiconductor Shortage

Tuesday, April 6th, 2021

A semiconductor shortage has been plaguing the automobile industry for several months, and this piece explains why:

To understand why the $450 billion semiconductor industry has lurched into crisis, a helpful place to start is a one-dollar part called a display driver.

Correction: The semiconductor industry itself isn’t in crisis, it’s making money hand-over-fist right now. It’s certain industries relying on semiconductors that have the problem.

Hundreds of different kinds of chips make up the global silicon industry, with the flashiest ones from Qualcomm Inc. and Intel Corp. going for $100 apiece to more than $1,000. Those run powerful computers or the shiny smartphone in your pocket. A display driver is mundane by contrast: Its sole purpose is to convey basic instructions for illuminating the screen on your phone, monitor or navigation system.

The trouble for the chip industry — and increasingly companies beyond tech, like automakers — is that there aren’t enough display drivers to go around. Firms that make them can’t keep up with surging demand so prices are spiking. That’s contributing to short supplies and increasing costs for liquid crystal display panels, essential components for making televisions and laptops, as well as cars, airplanes and high-end refrigerators.

“It’s not like you can just make do. If you have everything else, but you don’t have a display driver, then you can’t build your product,” says Stacy Rasgon, who covers the semiconductor industry for Sanford C. Bernstein.

Now the crunch in a handful of such seemingly insignificant parts — power management chips are also in short supply, for example — is cascading through the global economy. Automakers like Ford Motor Co., Nissan Motor Co. and Volkswagen AG have already scaled back production, leading to estimates for more than $60 billion in lost revenue for the industry this year.

A bit of background here: Back in the dim mists of time, some major car manufacturers used to have their own captive wafer fabrication plants for automotive components. They were more art-of-the-state than state-of-the-art, as well as heavily unionized. (Your etch machine broke? Better figure out whether you need the union plumber or the union electrician to fix it…) GM shut down their last semiconductor plan in Kokomo, Indiana (which I think was running a 500 nanomemter process, which was beyond old even then) in 2017.

The situation is likely to get worse before it gets better. A rare winter storm in Texas knocked out swaths of U.S. production. A fire at a key Japan factory will shut the facility for a month. Samsung Electronics Co. warned of a “serious imbalance” in the industry, while Taiwan Semiconductor Manufacturing Co. said it can’t keep up with demand despite running factories at more than 100% of capacity.

“I have never seen anything like this in the past 20 years since our company’s founding,” said Jordan Wu, co-founder and chief executive officer of Himax Technologies Co., a leading supplier of display drivers. “Every application is short of chips.”

The chip crunch was born out of an understandable miscalculation as the coronavirus pandemic hit last year. When Covid-19 began spreading from China to the rest of the world, many companies anticipated people would cut back as times got tough.

“I slashed all my projections. I was using the financial crisis as the model,” says Rasgon. “But demand was just really resilient.”

People stuck at home started buying technology — and then kept buying. They purchased better computers and bigger displays so they could work remotely. They got their kids new laptops for distance learning. They scooped up 4K televisions, game consoles, milk frothers, air fryers and immersion blenders to make life under quarantine more palatable. The pandemic turned into an extended Black Friday onlinepalooza.

Automakers were blindsided. They shut factories during the lockdown while demand crashed because no one could get to showrooms. They told suppliers to stop shipping components, including the chips that are increasingly essential for cars.

Then late last year, demand began to pick up. People wanted to get out and they didn’t want to use public transportation. Automakers reopened factories and went hat in hand to chipmakers like TSMC and Samsung. Their response? Back of the line. They couldn’t make chips fast enough for their still-loyal customers.

Here’s the crux of the problem:

Wu explained that he can’t make more display drivers by pushing his workforce harder. Himax designs display drivers and then has them manufactured at a foundry like TSMC or United Microelectronics Corp. His chips are made on what’s artfully called “mature node” technology, equipment at least a couple generations behind the cutting-edge processes. These machines etch lines in silicon at a width of 16 nanometers or more, compared with 5 nanometers for high-end chips.​

​The bottleneck is that these mature chip-making lines are running flat out. Wu says the pandemic drove such strong demand that manufacturing partners can’t make enough display drivers for all the panels that go into computers, televisions and game consoles — plus all the new products that companies are putting screens into, like refrigerators, smart thermometers and car-entertainment systems.

There’s been a particular squeeze in driver ICs for automotive systems because they’re usually made on 8-inch silicon wafers, rather than more advanced 12-inch wafers. Sumco Corp., one of the leading wafer manufacturers, reported production capacity for 8-inch equipment lines was about 5,000 wafers a month in 2020 — less than it was in 2017.

Hell, there are people still running some four inch fab lines out there, though usually it’s for something funky like gallium arsenide, old analog signal processes, etc.

The problem is, no one is building any new capacity in those old geometries because fabs are too expensive to build and need 2-3 years of lead time to get up and running. Moore’s second law states that the cost of a new, cutting edge semiconductor plant doubles every four years. You can’t just take an existing building and turn it into a fab, it has to be specially built from the ground up with exacting standards for cleanroom air filtering, concrete slab level uniformity, etc. And equipment manufacturers like Applied Materials and LAM Research aren’t going to sell you old technology machines to build older geometry chips because they’re not making them anymore. And if you have to pay full price for the equipment, you might as well fab higher-value chips in current geometries anyway.

TSMC is already spending $100 billion for expanded manufacturing capacity over the next three years, and Intel another $20 billion. That spiraling fab cost is why so many former integrated device manufacturers went to a fabless model, designing chips but letting the manufacturing be handled by foundries like TSMC, UMC and Global Foundries. (And Intel is expanding their own foundry business at the same time they’re paying TSMC to fab some of their top-end chips. You can’t tell the players without a scorecard…)

The other problem is the extremely cyclical nature of the semiconductor industry. In booms, fabs make money hand over fist. During busts, some segments (like RAM) barely break even. The foundry model has smoothed the spikes out somewhat, but as the current shortage shows, not entirely.

Just-In-Time delivery was one of the great disruptive business innovations. Leaner, more tightly-coupled computerized inventory lead to decreases in unused parts and faster times to market. But when there’s a hiccup in the supply chain, it makes it more immediately disruptive. It’s hard to obtain additional semiconductor parts if everyone’s fab is already at full capacity, so expect shortages to extend into the year.