Consumer Image Manipulation: Deep Fakes or Yes, That Is Granny!
September 7, 2022
I find deep fake services interesting. Good actors can create clever TikTok and YouTube videos. Bad actors can whip up a fake video résumé and chase a work from home job. There are other uses as well; for example, a zippy video professional can create a deep fake of a “star” who may be dead or just stubborn and generate a scene. Magic and maybe cheaper.
I read “Use This Free Tool to Restore Faces in Old Family Photos.” The main idea is that a crappy old photo with blurry faces can be made almost like “new.” The write up says:
This online tool—called GFPGAN—first made it onto our radar when it was featured in the August 28 edition of the (excellent) Recomendo newsletter, specifically, a post by Kevin Kelly. In it, he says that he uses this free program to restore his own old family photos, noting that it focuses solely on the faces of those pictured, and “works pretty well, sometimes perfectly, in color and black and white.”
The service has another trick amidst its zeros and ones:
According to the ByteXD post, in addition to fixing or restoring faces in old photos, you can also use GFPGAN to increase the resolution of the entire image. Plus, because the tool works using artificial intelligence, it can also come in handy if you need to fix AI art portraits. ByteXD provides instructions for both upscaling and improving the quality of AI art portraits, for people interested in those features.
Will it work on passport photos and other types of interesting documents? We will have to wait until the bad actors explore and innovate.
Stephen E Arnold, September 8, 2022
A Surprise: Newton Minnow Was Prescient
August 30, 2022
Social media is to blame for most misinformation spreading across the Internet faster than viral videos. Despite declining numbers, TV still plays a huge part in the polarization of the American populace. Ars Technica explains why: “It’s Just Not Social Media: Cable News Has Bigger Effect On Polarization.” While social media echo chambers exist, it is not at the huge scale we have been led to believe.
Researchers from Microsoft Researchers, Stanford University, and the University of Pennsylvania tracked TV consumption from thousands of American adults between 2016 to 2019. They discovered that selective news exposure did increase polarization, but it mostly came from TV. They found that 17% of American TV news watchers are politically polarized with a near-split average between left and right politics. That is three to four times higher than online news watchers.
TV watchers also do not change their viewing habits:
“Besides being more politically siloed on average, our research found that TV news consumers are much more likely than web consumers to maintain the same partisan news diets over time: after six months, left-leaning TV audiences are 10 times more likely to remain segregated than left-leaning online audiences, and right-leaning audiences are 4.5 times more likely than their online counterparts. While these figures may seem intimidating, it is important to keep in mind that even among TV viewers, about 70 percent of right-leaning viewers and about 80 percent of left-leaning viewers do switch their news diets within six months. To the extent that long-lasting echo chambers do exist, then, they include only about 4 percent of the population.”
Also depending on the TV viewers’ political leanings, they never stray too far from preferred news networks. The political imbalance is increasing among how audiences get their news, because more are shifting from broadcast news to cable.
This is not good, because it increases divisions among people rather than showing the commonalities everyone shares. It also makes news more sensational than it needs to be.
Whitney Grace, August 30, 2022
Quantum Supremacy Emulators: The Crypto Claim
August 16, 2022
I noted the silliness of the quantum supremacy claims first by the GOOG and then by the Red Hat dependent IBM. I pointed out that Intel claimed a quantum thing-a-ma-bob that would be a hub for certain quantum functions. Yeah, horse something, maybe ridge, maybe feathers. I mentioned in one of my blog posts or client emails that the US government aided by big wizards had developed algorithms that could not be broken by yet-to-be-invented quantum computers.
Now we have an interesting story that puts much of the quantum supremacy-type PR in a flaming dumpster. Wow, look at the dense smoke from a piddling fire.
“Post Quantum Encryption Contender Is Taken Out by Single-Core PC and 1 Hour” states:
SIKE is the second NIST-designated PQC candidate to be invalidated this year. In February, IBM post-doc researcher Ward Beullens published research that broke Rainbow, a cryptographic signature scheme with its security, according to Cryptomathic, “relying on the hardness of the problem of solving a large system of multivariate quadratic equations over a finite field.”
Everyone will keep trying. Perhaps a functioning quantum computer will become available to make hunting for flaws more helpful. No, wait a minute. The super algorithm was compromised by a single core PC chugging along for one hour.
Oh, well, as long as one doesn’t look too closely some of the quantum supremacy PR sounds great. In my opinion, some of the stuff is a bit silly.
Stephen E Arnold, August 16, 2022
Intel: Optane Offline
August 12, 2022
I group Optane in the Intel “horse feathers” category. There is another point of view, and I want to highlight because different ideas are useful. The none-horse feather angle is expressed in “Why the End of Optane Is Bad News for All IT. The Biggest New Idea in Computing for Half a Century Was Just Scrapped.” (I will not point out the “all” word. I will not remind you, gentle reader, that quantum computer is also one of the big ideas in computing in the last half century. I want to, but I will be restrained.)
The article romps through the history of no file systems, sort of file systems, clunky methods of moving zeros and ones to and fro, and related milestones. Here’s the main point of Intel Optane:
No more installing OSes, no more booting up. No more apps. The OS sits in memory all the time, and so do your apps. And if you have a terabyte or two of nonvolatile memory in your computer, what do you need SSDs for? It’s all just memory. One small section is fast and infinitely rewritable, but its contents disappear when the power goes. The other 95 per cent holds its contents forever.
I understand. I think that the technical idea was darned good. However, the flaw in the Intel method is stated clearly in the write up, just more delicately than my sweeping the Intel method into the pile of horse feathers I favor. Here’s the sentence I think nails it:
But Intel made it work, produced this stuff, put it on the market… and not enough people were interested, and now it is giving up, too.
“Giving up.” Intel has substituted finding a way to make it work for PR and marketing. With the CHIPS coming, Intel will have a chance to deliver, if not at Apple nanoscale.
What makes me nervous about technology outfits today is that “good enough” is now defined as “excellence.”
“Giving up” is working hard to make good business decisions. Intel must demonstrate that it can deliver old fashioned excellence and persistence. You know just not “giving up.”
Stephen E Arnold, August 121, 2022
Big Tech: Is Big Change Next to Impossible?
August 3, 2022
I read “The Disproportionate Influence of Early Tech Decisions.” The article adds some specificity to the notions of technical debt and why the Roman empire ended up with people wearing fur in the summer recycling stone from gigantic weird buildings jammed together. There are nifty quotes about the nature of things. (Lucretius and De Rerum Natura, right?)
The more recent article states:
… everybody knows that it’s hard to migrate a database or rewrite code in a new language, so this status quo wouldn’t be surprising anywhere you find it….What is more surprising is that it’s not only the big stuff that has a tendency to stay fixed. It’s the small and medium-sized elements as well.
I interpret these observations to the plight of the Silicon Valley type of big tech outfits. I believe that the observation applies to pinnacles of technological capability like the US air traffic control system and the Internal Revenue Service. If you have a backlog, just shred it. Effective. Simple. No big tech needed.
The article illustrates how expedient (maybe just bad?) initial decisions persist through time. There are examples of fixing and adding, but the persistence of initial conditions is a characteristic of some companies’ products and services.
The point which resonated with me was:
Simply this: software has inertia.
Bingo!
I noted these statements too:
quality is more of a sliding scale than it is a good or bad dichotomy, and I’d argue that many small companies optimize too much in favor of speed by trading away too much in terms of maintainability by shipping the first thing that was thrown at the wall. And this fails the other way too, where major believers in academic-level correctness agonize over details to such a degree that projects never ship, and sometimes never even start.
So what?
The people and time required to figure out how to implement meaningful technical change impose a significant cost. Cost translates into management’s need to kick the can down the road, change jobs, or ignore the mounting problems. Early decisions manifest themselves in systems whose problems cannot be addressed; management decisions which to an outsider appear to be downright wacky; and big companies struggling to escape their past.
I never “meta” a high tech outfit that I could not google or rely on a one day delivery that stretches to 10 days or more or an operating system unable to print a copy of this blog post.
Vulnerable? You bet.
Stephen E Arnold, August 3, 2022
Smart Technology: There Will Be Glitches
August 2, 2022
Here’s a simple and clear question: What do like best about smart hardware, software, and systems?
[a] Everything
[b] Everything and the dividends paid on my shares in a company
[c] Everything even when there is a trivial glitch which will be fixed promptly?
Pretty nifty too. No wonder I was asked to resign from some group creating tests for fresh, nimble, young minds.
Consider two examples of smartness.
The first has been reported but since the incident took place in Moscow, the story did not have traction. “7-Year-Old’s Finger Broken By Chess-playing Robot” includes a video. The main idea is easy to grasp: The smart chess playing robot experienced a Tesla moment. Instead of running into a barrier on Highway 101, the robot snapped the child’s finger. Why not mount a weapon on a friendly robot dog? Right. No problemo.
The second mini case is described in “Computer Glitches Harmed Nearly 150 Patients after Oracle Cerner System Go-Live.” The main takeaway from this write up is:
Computer errors following the go-live of a new Oracle Cerner electronic health records system harmed nearly 150 patients at a Washington hospital, as revealed during a hearing in the US.
Net net: Technology is wonderful, does no harm, and definitely will benefit mankind. Young children and sick people? Well, maybe, maybe not.
Stephen E Arnold, August 2, 2022
Horsefeathering: The Intel Arc of Optane
August 1, 2022
Intel’s announcement of the Horse Ridge quantum thing caught my attention in 2020. Then there was Horse Ridge II a year later. I jotted in my notebook containing high-tech confections the idea for putting giant water consuming semiconductor fabs in Arizona. The idea for Optane seems to have fizzled.
Has Intel has reached peak “horse.” I dub the new era Horsefeather Arcs. Intel has not matched the privacy oriented outfit Apple’s chips. How far apart are these puppies? Far. Furthermore, Intel has not been able to blast past AMD and nVidia. Is Intel the future of the resurgent and reinvigorated semiconductor manufacturing sector? Sure, sure. There a big chip bill that is going to make this trivial task come true. Will it be similar to remediating Flint’s water issue? No problem, of course.
I read “Intel Arc Graphics Cards Could Be in Serious Trouble – Will Team Blue Throw in the Towel?” and formulated the concept of the Horsefeathers Arc.
The write up says:
Arc Alchemist and Battlemage might get the ax entirely over unfixable hardware flaws that are kneecapping their performance, and it’s threatening to scrap Intel’s entire Arc discrete graphics line.
Who says “the ax”? The write up’s author, that’s who? If true, will Alchemist and Battlemage produce analyst melting revenue? Maybe not?
According to the cited article:
This report comes from YouTuber Moore’s Law is Dead, and it is a doozy, full of internal politics, bitterness, and recriminations at Intel over the company’s graphics unit being unable to actually deliver the discrete graphics cards that have been hyped up for more than a year.
Even if this report from the cited article and the YouTuber, the delays and the reports about performance for Intel Arc are causing my confidence meter to curve toward zero. Horsefeathering?
Stephen E Arnold, August 1, 2022
Microsoft: Excellence in Action
July 25, 2022
I wanted to print one page of text. I thought a copy of the cute story about the antics of Elon and Sergey might be nice to keep. My hunch is that some of the content might be disappeared or be tough to see through the cloud of legal eagles responding to the interesting story. Sorry.
Nope.
Why?
Microsoft seems to be unable to update Windows without rendering a simple function. Was I alone in experiencing this demonstration of excellence? Nope. “Microsoft Warns That New Windows Updates May Break Printing.” The article states:
Microsoft said that the temporary fix has now been disabled by this week’s optional preview updates on Windows Server 2019 systems. This change will lead to printing and scanning failures in Windows environments with non-compliant devices.
There you go. Non compliant.
But wait, there’s more.
But wait there’s more!
“New Windows 11 Update Breaks the Start Menu Because Microsoft Hates Us All” explains:
It looks like Microsoft has once again shipped dodgy Windows 11 updates, with reports suggesting that the two latest cumulative updates have been causing serious issues with the Start menu. The updates in question are KB5015882 and KB5015814, and it looks like they’ve introduced a bug which causes to Start menu to disappear when you click to open it.
What do these examples suggest to me?
- A breakdown in basic quality control. Perhaps the company is involved in addressing layoffs, knock on effects from SolarWinds, and giving speeches about employee issues
- Alleged monopolies lack the management tools to deliver products and services which function like the marketing collateral asserts
- Employees follow misguided rules; for example, the Wall Street Journal’s assertion that employees should “ditch office chores that don’t help you get ahead.” See Page A 11, July 25, 2022. (If an employee is not as informed as a project lead or manager, how can the uninformed make a judgment about what is and what is not significant? This line of wacko reasoning allows companies with IBM type thinking to provide quantum safe algorithms BEFORE there are quantum computers which can break known encryption keys. Yep, the US government buys into this type of “logic” as well. Hello, NIST? Are you there.
Plus, Microsoft Teams, which is not exactly the most stable software on my Mac Mini, is going to get more exciting features. “Microsoft Is Launching a Facebook Rip-Off Inside Teams.” This article reports:
Microsoft is now launching Viva Engage today, a new Facebook-like app inside Teams that encourages social networking at work. Viva Engage builds on some of the strengths of Yammer, promoting digital communities, conversations, and self-expression in the workplace. While Yammer often feels like an extension of SharePoint and Office, Viva Engage looks like a Facebook replica. It includes a storylines section, which is effectively your Facebook news feed, featuring conversational posts, videos, images, and more. It looks and feels just like Facebook, and it’s clearly designed to feel similar so employees will use it to share news or even personal interests.
That’s exactly what I don’t want when “working.” The idea for me is to get a project, finish it, and move on to another project. Sound like kindergarten? Well, I listened to Mrs. Fenton. Perhaps some did not heed basic tips about generating useful outputs. Yeah, Teams with features added when the service does not do the job on some Macs. Great work from the Windows Phone and Surface units’ employer.
Net net: Problems? Yes. Fixable? I have yet to see proof that Microsoft can remediate its numerous technical potholes. Remember that Microsoft asserted that Russia organized 1,000 programmers to make Microsoft’s security issues more severe. In my view, Russia has demonstrated its inability to organize tanks, let alone complex coordinated software exploits. Come on, Microsoft.
Printers!
Stephen E Arnold, July 25, 2022
The Fix for Addiction Is More Addiction. Does That Sound Like a Solipsism?
July 21, 2022
I read a remarkable article on the Internet. I believe everything I read on the Internet. What could go wrong with that? (Hang on because one more philosophical puzzle awaits.)
“Samsung Says the Only Cure for Tech Dependency Is More Tech Dependency” caught my attention for two reasons.
First, the use of the word “only.” I like that type of categorical affirmative. It is just so positive in a world of intellectual gray. (Does Samsung make products tinted intellectual gray?)
Second, the solipsistic the fix for dependence is more dependence. Oh, yeah. That’s outstanding thinking.
The write up says:
You’re hooked on tech, and you wish you could stop. But wait, all you need is more tech, says Samsung — and its collaborator, Google.
Here’s another brilliantly sparkly gem:
Surely, though, we need a little more technology. Of course, we do.
This fentanyl-esque argument encapsulates the world view of the top quartile of the top one percent who work for Samsung and Google. Which is more representative of today’s technology environment? Samsung and Google teaming up to make tech the fix for everyone’s issues? The article about this outstanding idea? My pointing out the categorical affirmative and the solipsistic rhetoric? Blame me. Go on. It’s okay. I have a new high tech cable which neutralizes your criticism.
Stephen E Arnold, July 21, 2022
How to Point Out a Consulting Outfit Is Often Full of Beans
July 19, 2022
I read a write up in the UK online publication The Register. The article was “IT Departments Often Regret Technology Buying Decisions.” I immediately thought about Google’s mantra that organizations did not need information technology departments. I think the reasoning behind the statement was, “Let Google do it because we are smarter and have scaling, analytics, smart software, etc., etc.” I first heard this mantra in the 2002, maybe 2003 period. I wondered if the article was just recycling Google-type fluff-a-roo?
Yes because I have heard this before. Nope because the mid tier consulting firm is probably unaware of the world before checking TikTok in the last 10 minutes.
The write up pivots on a mid tier consulting firm which has “reinvented” the Google-type mantra with a bit of the rap music beat.
I learned:
Fifty-six percent of organizations said they had a high degree of regret over their largest tech-related purchase in the last two years, according to a new survey of 1,120 executives in North America, Western Europe, and Asia/Pacific.
Ok, almost 60 percent are faced with a persistent problem. This is not technical debt; this is here-and-now craziness.
I found this passage a slightly nicer way of saying what the Google-type mantra arrogantly implied:
… For anyone left picking up the technical pieces, 67 percent of people involved in technology-buying decisions are not in IT, which means that anyone could be a tech buyer for their organization. This is the so-called lines of business phenomenon where someone in marketing, for example, uses the corporate credit card to buy a product or service that IT admins then have to help manage.
Who is best qualified to make technology decisions for an organization? The answer is obvious:
- MBAs who can use Excel
- Accountants who can use a pencil and paper
- Lawyers who can use Word and maybe a time reporting system
- Marketing professionals who can use gym equipment, acrylic paints, and art museum audio tour gear.
The outfit creating this report is a mid-tier consulting firm.
Now here’s the way to put the obvious into a for fee report:
Whether anyone has experienced buyer’s remorse after shelling out thousands of dollars for a Gartner report is a question upon which The Register cannot comment.
Bingo. Very obvious report. An expensive mid tier report which could have been summarized by talking to a Googler more than a decade or more ago. And the remarkable inability of experts to perceive that their expertise is a reflection of the present technology environment. Score: Mid tier zero. Register one.
Stephen E Arnold, July 19, 2022
-
- Subscribe to Beyond Search
Feature archive
News archive
-
