HNHacker News
TopNewBestAskShowJobs

ygjb

2,394 karma · joined September 8, 2011

Yvan Boily Founding Security Engineer Co-founder, Oghmere parentsguidetoai.ca https://nimblemyth.ca
submissionscomments
ygjb··on OpenAI acquires Global Illumination
Maybe not, but frequently hiring a team that has accomplished a significant percentage of a goal is a better choice than trying to retask existing folks, or to grow a new product set outside of core competency.
ygjb··on Nintendo is trying to patent some broad Tears of the Kingdom mechanics
My kids are both actively playing TOTK on their profiles on my switch with a single digital purchase.

There have been other games where this has been problematic (Animal crossing I think limited some features per device instead of per account).

ygjb··on A room-temperature superconductor? New developments
I think it's more likely that the bulk of folks don't fully comprehend why this would be so important. I would imagine alot of reporters look at the things they could write about and the choices are politics, war, polarizing news, or nerds nerding hard trying to prove other nerds right or wrong, and it just doesn't make the cut.
ygjb··on Amazon requires services on Fire TV to give it 30% of ad impressions or revenue
It's not really misinformation though. Most vuln researchers aren't burning browser 0-days targetting pirates, but many piracy sites host malicious content, even if that content isn't looking to deliver malware through browser vulnerabilities.

There is a significant segment of the internet that thrives in the gutters between paid services, straight up piracy sites, and actual "dark web" sites that require TOR or other specialized access. The restreamers that live in those gutters are questionable at best, straight up illegal at worst, and fund themselves through fraudulent ads, or legit ads for sketchy or dangerous services and sites.

It is entirely possible to use those sites safely, just like it's also possible to navigate dark alleys in major urban centres in the late hours, but that doesn't make it a low-risk activity.

ygjb··on Amazon requires services on Fire TV to give it 30% of ad impressions or revenue
It's less an issue of the reality distortion field, and more the issue that Apple hasn't been co-opted by the allure of advertising dollars. That's going to change, with projections of revenue growth just from advertising expected to go from $5.3B to ~$20B (industry expectations), and regulations like the Digital Markets Act explicitly going after their walled garden.
ygjb··on The New York Times attacks e-bikes while ignoring the real danger all around us
Vancouver, a very bicycle friendly city, has numerous shared pedestrian and bike paths. Most of the ones that are shared are usually sidewalk transitions between bike lanes and busy streets where there simply isn't space to build a separated bike path.

That said, many, many of the bike and pedestrian paths are adjacent and are only separated by a painted lines, and signs urging pedestrians to stay out of bike lanes :/

ygjb··on Betterfox: User.js to harden Firefox and optimize privacy, security, and speed
They do modify the attack surface, but since these prefs cover code that ships in the release version of Firefox, they should be reasonably well tested.

Most of the preferences in this will actually reduce the target-able attack surface, and disable some features that could be footguns, but at the same time disable some optimizations that aim to reduce cognitive load on users (which in turns reduces the likelihood that mistakes will be made).

ygjb··on Betterfox: User.js to harden Firefox and optimize privacy, security, and speed
There is a crucial difference between a config file that changes options within Firefox, and something that adds new code.

Unless something has changed dramatically, the Firefox team (and other teams) test and fuzz Firefox extensively - obviously default prefs and features will get more coverage.

An actual fork like LibreWolf that adds or removes code and ships a different build fundamentally changes how the software is built and invalidates a significant portion of the security work that has been done by the Firefox security team.

I am not arguing that Librewolf or any of the other forks of Firefox[1] are necessarily worse for security, just that browsers are ridiculously complex pieces of software and making claims about security requires more than just gutting or adding features.

[1] other than Palemoon, because screw those guys; when I challenged them on the security program for Palemoon, their loopy leader tried to harass me and other former Mozilla folks. Given their propensity for instability I wouldn't trust them at all.

ygjb··on Global heat in ‘uncharted territory’: 2023 could be the hottest year on record
https://www.reuters.com/graphics/GLOBAL-ENVIRONMENT/INSECT-A...

https://www.scientificamerican.com/article/scientists-warn-o...

Just two quick examples, but go ahead and keep your head in the sand.

ygjb··on Global heat in ‘uncharted territory’: 2023 could be the hottest year on record
There is nothing that individual consumers can do that will change this. Policy is needed at a national and international level.
ygjb··on Global heat in ‘uncharted territory’: 2023 could be the hottest year on record
Right - global warming as a function of climate change, it's not the climate that is collapsing, it's the ecology that ~8 billion people depend on to survive that is collapsing :)

fiddles pedantically while the world burns

ygjb··on TeleSign profiles half of the world’s mobile phone users
> Most of the online banks are pretty secure for non-oblivious person.

Ok, granted, I spent the last 23 years of my life working in IT security across consulting, government, finance, and tech companies, but this is just garbage. Banks only invest in security to the degree that: - they are legally required to - they have contractual obligations to - that the risk of loss for a specific class of incident exceeds their self-insurance threshold

That's not a hypothetical comment, that is something that was explained to me as an AppSec lead when running into walls trying to get some issues fixed at one of the largest banks in the world. For the record, the issues that I was trying to have remediated would have had to exceeded an annualized loss expectancy for the region I was operating in of 10 million dollars per year to be considered risky.

Your definition of a bank being pretty secure and mine are probably radically different.

> Combination of user & password with enough entropy, and basic brute-force defense that blocks after 3-4 attempts is the industry minimum standard.

Sure, users should choose strong passwords. Banks should also require multi-factor authentication (real 2fa, not the SMS based weaksauce that a bunch use). But, that increases support and transaction costs. So, instead, blame the user! Beyond password selection, there is also the issue of how passwords are hashed, salted, stored, and brokered into a more reliable back-end credential that can be used, absolutely none of which the user has input into or control over, but sure, blame the user.

> User is the weakest link always, you cannot fix the "stupid" user that downloads malware, warez, adult content and gets infected and loses everything.

sigh you really like banging that drum.

> These people need life lesson to learn how to operate technology safely. > Although I agree that online banking could be made more secure, but the threat model will immediately evolve and adapt because scammers/fraudsters are still there and they want to eat.

There is absolutely no way to train average users to operate modern internet technologies safely because the average user has no effective control over the software and hardware they use (yes, Linux is a thing, and so is open source hardware, but users of those OS and hardware are not average users)

The primary reason the incidence of fraud is so high in the finance sector is because business has chosen to optimize for high transaction volume, and has accepted the risks of doing so. Stop trying to blame end users.

ygjb··on TeleSign profiles half of the world’s mobile phone users
> We require drivers license to operate vehicle, it is time we should require infosec101 training before handing over credit cards and or online banking accounts.

Sure. Why not start with an outline for what infosec101 should look like. Include estimates for how long the training should take, what the cadence for testing should be, and which agency should be responsible for validating that training. Do be sure to accurately communicate the degree to which an end user with a chip enabled bank or credit card has the ability to distinguish and disambiguate what constitutes a 'safe' or 'legitimate' online business. Also, include some details about how individuals who have been certified as completing this class and/or licensing scheme should procure insurance to protect themselves in case of an accidental data breach (for example, they leak their card info), and outline the process by which that same licensee can file an insurance claim against the insured party downstream of the physical point of payment or online payment portal that allowed a breach to happen. After all - if we are going to require online safety training, and licensing, then we should create another insurance scheme to facilitate resolution of those claims and resolve the costs.

It is really easy to point the fingers at a customer and say "problem exists between chair and keyboard", but the reality is that in the modern economy, the end user has almost no control over the security of their transactions, and little ability to influence how their purchase is handled beyond the question of "cash or card".

The only incentive that retailers, online stores, payment processors, and financial institutions have to resolve this is the simple fact that they own the liability for this, and it's only through the myth of the idiot user that they have been able to shift that liability, to varying degrees, back to the consumer.

ygjb··on TeleSign profiles half of the world’s mobile phone users
So point by point: 1. Stealing a credit card number, ccv, expiry, and name shouldn't be game over. It is because businesses want to reduce friction for customer purchases. If card not present transactions required a second step of verification, for example, multi-factor authentication, it would drastically reduce this type of fraud. Unfortunately it would also increase friction for online purchases, which is why everyone in the card processing chain is looking to shift liability away from themselves.

2. That is not the customers fault. Full stop. Yes, some sites are more shady than others, but there is nothing a consumer can do to determine if a service provider will get hacked.

3. Yes. Unfortunately, phishing is really easy. Despite the prevalence of this attack, training users to effectively detect and avoid being a victim is almost impossible.

4. See #3.

5. See #2.

6. How is a customer supposed to validate the security of an ATM against modern skimming technology, many of which are virtually indistinguishable from normal bank machines.

7. Yep, not great. Why don't banks require 2FA? Because it creates friction and increases costs. Better to just externalize the risk.

Your entire blame the user argument is bunk that has been packaged up and recirculated by the finance community for almost 20 years (and I have been using these arguments against them for nearly that long, granted it's close to ~12 years since I worked in infosec at a bank).

ygjb··on The first 7 potential “Gatekeepers” under the EU Digital Markets Act
I think it's really interesting that Sony is not on that list. It is self-declared, but Sony had > 75B€ in revenue in EU in 2022, I would guess has more than 10k business users across it's lines of business, and guessing from the sales of PS4 and PS5 units alone must be close, if not over the monthly active users threshold.

All things considered, Nintendo has a bit of a climb to get to those numbers.

ygjb··on Tell HN: People forget that you can stick any data at the end of a bash script
Sure, but that's turtles all the way down... any time you run untrusted code, you are making a risk based decision, usually based on the provenance of the code.
ygjb··on Mozilla restricts extensions on some domains on Firefox 115
Nah, it was meant as preachy, but not necessarily condescending.

It's absolutely important to challenge Mozilla and other open source projects, especially in this era of enshittification[1]; Mozilla and Firefox operate in a position of trust on behalf of their users.

That said, the parent post positioned this as an abomination of a feature, but acknowledged it makes sense as a user feature. The ability to disable add-ons by domain is a great feature for user control, but it's functionally useless on it's own as a mechanism to protect users.

In order for that feature to actually protect users, you need a mechanism to turn it on and off remotely so that if a new threat is identified (or there is a serious regression in Firefox that makes specific extensions higher risk), that users don't need to act to do the right thing.

This isn't a meaningful loss of user control, and I already said elsewhere that Mozilla should have communicated more about this new feature, but ultimately it's the right kind of feature.

[1] https://pluralistic.net/tag/enshittification/

ygjb··on Mozilla restricts extensions on some domains on Firefox 115
> Preventing the random extension I installed from hijacking my bank login page is good! Giving Mozilla the ability to disable my adblocker or NoScript on an arbitrary domain list that they can update remotely is scary!

So the ability for the web browser to arbitrarily add and remove features from the browser is scary? Just asking because there is a massive security trade-off and the intersection of a number of threat models in this comment.

Do you trust the platform you use to download and execute arbitrary code (that is, web content) to automatically update itself?

If not, how do you balance the lack of automated updates against the need to keep software up to date to prevent exploit of known vulnerabilities?

If so, how do you distinguish the ability to download and execute new code that could remove or suppress the features you choose from the ability to enable and disable add-ons/extensions?

There could have been better communication on this, but describing the feature as scary tells me you don't really understand the threat model around your use of a web browser, and may not be asking the right questions or considering the actual threats.

ygjb··on Mozilla restricts extensions on some domains on Firefox 115
It's actually ok for you to feel that way! It's also ok for Mozilla to do this, because Mozilla aims to use this to protect users! The internet is already a yard full of rakes for folks, I appreciate things that make it easier for users to protect themselves online.

Yes, the feature can be abused, but frankly, at least Firefox is an open source project, and there are methods that can be used to disable this feature, up to and including using or creating a new Firefox fork.

ygjb··on Sao Paulo: A city with no outdoor advertisements (2013)
On point 3, well, yes. If your mother or teacher wanted to, and had the means and opportunity to do so. Whether or not you find artistic merit in those drawings has little bearing on the expression of pride that might drive your parent or teacher to publish that art. It doesn't have any bearing on the feelings that are evoked by the observer, whether it's your revulsion at your own productions, or the delight from normal folks when seeing displays of public art.

You said "Real art is deep, meaningful, timeless and doesn't need to be imposed."

Sand Mandalas are deeply meaningful, but completely timeless - most are created to be destroyed, often within days or hours of completion. Brueghel's Flatterers is a relatively crass, surface level, but lasting commentary on what we often refer to as Musk fanboys and such these days :P

Fundamentally, art is human expression, and your attempts to gatekeep that have fallen pretty flat so far.

ygjb··on How Dowries Are Fuelling a Femicide Epidemic
I think a lot of folks who don't have an agricultural background mistake how different the processes are. Often when a person uses a hoe in modern times (for example, urban gardeners), they use it the way one would use a plough - long troughs or lines, except that they then plant seeds in a a few locations on the line, with the belief that turning the earth has some significant impact. Tilling can be a valuable process depending on what you are growing, the quality of the soil, when it is done, etc, but it is very distinct from using a hoe to plant seeds.

In practice, planting with a hoe is extremely light work, you turn the earth (or even just make a little hole), plant the seeds, maybe apply some fertilizer, or fresh soil, then close up the hole.

With a plough, you cut a long line, usually with pack animal or mechanical assistance, and then follow and plant seeds along the entire line. This can mean managing a large animal, or operating heavy machinery, or both! Plus there is the issue of removing obstacles (dead trees/plants, rocks, and other debris).

Setting aside the physical requirements for the different planting methods, there is also the issue of the capital cost of farming. https://blog.oup.com/2013/06/agriculture-gender-roles-norms-... is a decent high level article that covers how and why this is thought to contribute to gender roles.

ygjb··on AI camera with no lens
Police in your jurisdiction are held accountable?
ygjb··on Supreme Court rules Andy Warhol’s Prince art is copyright infringement
so.... no intergenerational wealth transfer then? if estates can't pass on intellectual property, then why should they be allowed to pass on physical property? If they can pass on physical property, then why not intellectual property, at least within the legally defined (and IMO far too long) windows of protection?
ygjb··on Sam Altman goes before US Congress to propose licenses for building AI
I mean, it's in the headlines regularly, but sure, I'll google it for you.

https://www.scientificamerican.com/article/racial-bias-found...

https://www.propublica.org/article/machine-bias-risk-assessm...

https://www.theguardian.com/technology/2018/oct/10/amazon-hi...

Three easy to find examples. There is no shortage of discussion of these issues, and they are not new. Bias in new technologies has been a long-standing issue, and garbage in, garbage out has been a well understood problem for generations.

LLMs are pretty cool, and will enable a whole new set of tools. AI presents alot of opportunities, but the risks are significant, and I am not overly worried about a skynet or gray goo scenario. Before we worry about those, we need to worry about the bias being built into automated systems that will decide who gets bail, who gets social benefits, which communities get resources allocated, how our family, friends, and communities are targeted by businesses, etc.

ygjb··on Sam Altman goes before US Congress to propose licenses for building AI
Yes, guns don't kill people, people kill people.

We know, we have watched this argument unfold in the United States over the last 100 years. It sure does seem like a few people are using guns to kill a lot of people.

The point of regulating AI should be to explicitly require every single use of AI and machine learning to be clearly labelled so that when people seek remedy for the injustices that are already being perpetrated, that it is clearly understood that the people who chose to use not just ML or AI technology, but those specific models and training criteria can be held accountable if they should be.

Regulation doesn't have to be a ban, or limits on how it can be used, it can simply be a requirement for clearly marked disclosure. It could also include clear regulations for lawful access to the underlying math, training data, and intended use cases, and with financially significant penalties for non-compliance to discourage companies from treating it as a cost of doing business.

ygjb··on Some employers have decided to build their own housing for workers
Do you have a viable suggestion for an alternative provider?
ygjb··on Ask HN: Tell me about how you bombed a technical interview
I interviewed at Stripe, pre-COVID, and it was an onsite interview after I suffered a concussion. I should have cancelled the interview, but I was excited about the opportunity, and had underestimated how disabled the concussion left me.

Objectively I think I interviewed fairly strongly, however for the hands on coding portion of the interview, I basically couldn't even write more than a few lines. I couldn't remember simple things like function or class definitions. I had to look up a few syntax things, and my already challenging dyslexia was exacerbated to the point where I could barely type.

I recognize looking back on it that I had a freaking concussion and wasn't even supposed to be looking at screens or reading, but it was still really embarrassing.

ygjb··on Iceland long term visa for remote workers
I don't know about other countries, but immigration targets in Canada are trending towards attracting 500,000 immigrants per year by 2025. Just because we have alot of openings doesn't mean it's easy though... there are several paths, and most of them are designed to favour folks who will integrate well socially and economically (and that doesn't just mean rich white folks, it means more along the lines of skilled labour and the means to sustain your family while they get settled).
ygjb··on Today's 5-year-olds will likely live to 100
Well yeah, but the long tail of shitty health policy is coming home to roost. What kind of a health care impact do you think the recent criminalization of routine women's medical health procedures will have? (Hint - there are already women who are dying or having significant health impacts due to doctors delaying or outright refusing treatment out of fear of legal action in several states).
ygjb··on Today's 5-year-olds will likely live to 100
Those are all individuals who had access to the finest health care on the planet, and teams of folks at various periods of their lives to focus on them and their health. Access to health care, and personalized health care on top of that is just one other aspect of the degree of luck it took them to reach those ages.

Whether it's a car accident, an earthquake, or an aggressive oncogene that gets activated by a cosmic ray or cup of diet pop, there are way too many things that will straight up kill us despite our otherwise healthy or privileged lifestyles.

Treating diseases does have an impact on extending the average lifespan, but early health interventions (that is, vaccinations, fewer babies dying from preventable illness or malnutrition, and women's pre- and peri-natal health care) has been far more impactful. Stretching that long-tail out in the future is going to require significant advances in treating and remediating aging and it's related indignities.

← PreviousPage 11 of 24Next →