Hacker Newsnew | past | comments | ask | show | jobs | submit | more BiteCode_dev's commentslogin

> We can't pretend that something essential for the process doesn't matter because we do a classification handwaving. It all counts

Funny thing: in France, stats for wind energy include the cost of wiring, and cables are required to be buried. Stats for nuclear just ignore the wiring. I'm pro nuclear, but that always bugged me.


Google analytics won't track cross site ?


Does Google Analytics even track cross-site today? Looking at it in developer tools I only see it using first-party cookies.

But anyway, Google Analytics won't be able to do it because nobody will be able to do it. For example, here is Chrome's project to remove cross-site tracking: https://www.chromium.org/Home/chromium-privacy/privacy-sandb...

(Still speaking only for myself)


I'm not sure if you're being intentionally obtuse or not. GA phones home with vast information about user, and builds a profile of them. That profile is correlated across sites to personalize search results and sell ads.


Please omit personal swipes from your HN comments. Your post would be fine without the first sentence.

Note this site guideline, including the last bit: "Please respond to the strongest plausible interpretation of what someone says, not a weaker one that's easier to criticize. Assume good faith."

https://news.ycombinator.com/newsguidelines.html


I'm not being intentionally obtuse, but I also don't know all of Google's advertising business. I didn't think Google Analytics did that? What makes you think it does?

(GA sends a message to Google, but I had thought that it was not linked to your behavior on other sites via GA?)


google analytics tells you the interest of your audience. how do you think it does this without correlating you to a profile they've built?


Googe Analytics consultant here (not Google employee).

1. Google Analytics' primary identity signal is a first-party cookie. this is not shared between domains. There is no technical way to link identity between domains with different cookie values.

1a. Google Analytics has built-in library functions to allow site owners to share first-party cookie values between a whitelisted set of domains. This effectively lets one company with multiple sites share a first-party identifier, but still not let anyone (Google or otherwise) link that identity to identities set on other sites.

1b. BUT. But. BUT. Google is rolling out "Google Signals" for Google Analytics, which will use your Google Account as the identity signal instead for users who are logged in to Chrome. This, obviously, lets your identity be correlated across sites.

(Personally, I suspect that the availability of this feature played a part in Google's decision to let Chrome follow the industry towards blocking third-party cookies. But this is a baseless opinion, one step removed from a conspiracy theory.)

2. Google Analytics can link their identifier (the first-party cookie or Google Signals) to your DoubleClick profile via DoubleClick's third-party cookie. The checkbox that does this is unchecked by default. There are many other features of GA that encourage or require you to check this checkbox.

2a. Google's documentation (including legal contracts!) places limits in the data exchanged between the two profiles. Data exchanged does include demographic and interest information from DoubleClick's profile into GA. This is one of the big reasons why people click the checkbox.

To my knowledge, GA data is not used to inform the DoubleClick profile. GA data can be used to build an "audience" in various Google ad platforms, and direct ads to those people specifically, or to use as the basis for a "look-alike audience."

3. Google is a Processor under GDPR for Google Analytics, and a Controller under GDPR for Google Ads. To a first approximation, this means they make the specific legal claim that they do not use GA data for their own purposes. Linking Analytics and Ads data is... complicated and frankly I still haven't gotten an explanation of its legal status that I fully understand.

In my personal opinion, I don't think Google actually uses Google Analytics data. Most Analytics implementations are tire fires, and they can get all the data from other more reliable sources, like Publisher data or Chrome. Given that they have based on their entire GDPR compliance strategy for Analytics on being a Processor, I don't think the risk/reward is there.

(apologies for lack of copy-editing, the thunder's about to take my internet away)


Interesting insight. Their privacy policy suggests they are able to do this however[1]:

"When you’re not signed in to a Google Account, we store the information we collect with unique identifiers tied to the browser, application, or device you’re using."

What is stopping, legally, them from taking e.g. HTTP Headers from independent connections and linking them together through fingerprinting? Maybe this is not implemented in Google Analytics, but that is certainly not the only connection made to Google on most websites (see e.g. gstatic and Firebase). Since there is practically no technical barrier, it seems that the vague privacy policy leads to the only question being what the 'unique identifiers' are exactly.

[1] https://policies.google.com/privacy?hl=en#infocollect


> What is stopping, legally, them from taking e.g. HTTP Headers from independent connections and linking them together through fingerprinting?

Setting aside legality, the attack you're describing will be thwarted by network state partitioning: https://www.chromestatus.com/feature/6713488334389248


Am I missing something, or does this not address the fact that https://whatsmybrowser.info/ will always yield the same thing whether a connection is made to GA, Firebase, YouTube or Blogger? That seems to have to do with connection timing and caching, not with what the connection itself leaks.


I thought the parent was talking about using information about the network connection to link users across sites hosted by the same entity?

If you're talking about fingerprinting in general, that is also something that all the browsers are working on. I'm most familiar with Chrome's strategy, which is to first switch APIs that provide a lot of entropy from something you get by default to something you have to actively request, figure out how to provide similar functionality more privately, and then enforce a privacy budget that does not allow collecting enough information to identify users: https://www.chromium.org/Home/chromium-privacy/privacy-sandb...


Is Google going to give back the $100B is made from cross site tracking in the past?

Is Google going to consider YouTube, Gmail, Maps, and Android Location history different sites, or is "having an effective monopoly an exemption to crosss-site tracking prohibition?

Does anything in the proposal prevent server-side cross-site tracking? (No.)

Is Google going to stop buying third party tracking data like credit card transactions?


This is BS, it states 3 cents for France, but most people have unlimited plans for 20 euros here.


That's because:

- bad trips are rare if your respect set and settings, which every responsible consumer will tell you to do.

- they have been used politically in bad faith. It's only fair to not add to the stack of propaganda.

- most of bad trips are not a bad thing. It can be spectacular, but rarely dangerous, and may actually be a good thing. My only bad trip helped me grow just like my good ones. It was not pleasant, just like throwing up stale food is not, but I was mentally healthier after it. Some dark things are inside of us all, and it's good they come out, but the experience is likely not something we crave for.


One should not simplify the work of a group people at the price of the entire community.


> I think splitting the market will be fine. However, I suspect many side-loading-capable iPhones are going to have many more security problems and privacy breaches.

Security and flexibility are always at odd with each other. But for side loading, I'm not really worried.

Users that are susceptible to security problems don't have the technical know-how to even do something as simple as sideloading. Those who are technical enough usually are not the entry point for malware.


This is not true.

Thanks to friendly YouTube and web tutorials, non- technical users are great at hacking themselves:

https://en.m.wikipedia.org/wiki/Self-XSS


>Users that are susceptible to security problems don't have the technical know-how to even do something as simple as sideloading. Those who are technical enough usually are not the entry point for malware.

If sideloading is made too difficult for average users, it would totally negate the anti-trust (competitive) benefits of forcing the platform open in the first place.

Either governments make sideloading easy and approachable to the masses, and accept the UX/security/privacy risks involved, or they make it cumbersome, and accept the continuation of the App Store monopoly.


I'm not sure about that. I do see a lot more intellectual rigor in the older techies than in the ones from my generation. Just basic grammar illustrate that well.

And it makes sense, if you can't rely on automatic systems to do part of your work, you will be trained by your daily tasks to be a more powerful thinker.

If what the company do is hard, being ok will not cut it. You cannot google your way into innovation, you can't copy/paste architecture design, and your calculator won't save you from a logical mistake.

I'm personnally very adapted to agile envs with margin of errors and a lot of feedback loops. But a waterfall is more challenging, because I'm not born in it. And you don't use scrum to build the path to moon landing.

More than that, I arised in the "a good dev is lazy" period, where working smart, not hard was praised. But hitting 30, reality calls back: there are no shortcut to awesomeness, you will have to work hard. And not many engineers are ready to do so. The ones who do often create their company.


> I do see a lot more intellectual rigor in the older techies than in the ones from my generation.

My experience has been the opposite - though as this is based on observation it's obviously subjective - and opportunities for selection-bias are also present (i.e. how many of the last-generation that didn't get promoted-out of being a technical contributor was because they were more valuable to the company that way?)

Your remark about automated tooling is interesting - because I feel that modern tooling (TLA+, Z3, constraint-provers, and cutting-edge languages like Haskell, Idris, and so on) really do require almost a postgraduate-level of understanding of the CS theory involved - whereas if you look at _the SE scene_ in the 1970s and 1980s - or even the mid-1990s, the tooling certainly did require you to do more planning and reasoning ahead of time (VB6, lol) but I can't see the entire industry of the time doing their modelling and verification entirely by themselves: on the contrary (and based on the horror-inducing programming code I've seen) a lot of it was ad-hoc and trial-and-error - Visual Studio didn't get built-in support for unit-tests until 2008 (or 2005 if you had the expensive edition). Also consider that the old SE processes used back then (Waterfall, boxed software, slow-moving-and-big release cycles, etc) meant there was more room for less-rigorous folks in large software dev teams.

> and your calculator won't save you from a logical listake.

For that, you need a spell-checker!

UPDATE: Ah, you edited your post, which ruins the joke :/


With all due respect, I think you are mixing apples and oranges here. "TLA+, Z3, constraint-provers, and cutting-edge languages like Haskell, Idris, and so on" is not exactly the bread and butter of IT work nowadays, so why do you think it is appropriate to compare it with "VB6, lol"?


(When you say "IT work" - are you referring to software-engineering work specifically - or "IT" in general? (I'm personally not a fan of "IT" as a term at all because it's unhelpfully vague[1]).

I wasn't comparing them directly like that.

My point is that in the 1990s if you used VB in production systems then you would effectively be forced into investing a lot of company resources - and mental effort - into all kinds of entirely disconnected approaches for formalising, proving and even testing your system. Whereas the state of the art for formalising SWE work today currently lies in tools like TLA+, Z3, and others and how it's significantly easier to map those formal models to our production program code back-and-forth (when written in more expressive and formal-friendly languages like Haskell, etc) than it was 20+ years ago.

Back then the sheer effort involved to perform even automated unit testing and integration testing of VB6 code, and other languages of a similar nature, was massive because, not least, the VB6 tooling completely lacked that functionality, and their lack of terse expressivity meant time spent just writing repetitive code with little value - and things were tenfold worse if you were using one of the myriad proprietary other 4GL languages of the time because most of those vendors swore-off any kind of interoperability or extension mechanism because they saw them as a threat to their business model (as a textbook example see Progress' utterly laughable defense of maintaining their position here: https://www.progress.com/tutorials/odbc/open-source-database despite their market-share shrinking and the company quickly pivoting away from their own database system and onto being a generic component/tooling vendor).

Even so, I'm not being ageist: VB6 wasn't state-of-the-art when it came out: Java is older and far better in that regard (VB6's type system is very anemic). I was singling out VB6 for criticism specifically because even back then it wasn't very good, it's just disappointing that so many people used it as the basis for production systems.

[1] Whenever UK friends and relatives refer to me as "working in IT in the 'states" I outright deny it: I really don't feel that I work in "IT".


Whoa, a fellow PROGRESS user?!

Anyway, thanks for the answer. I am not sure I completely agree with you, but at least now your point is more clear.


I’m not a Progress user. Instead I get paid lots of money to help companies move-away from Progress to Postgres :)

(Which in practice just means figuring out how to get their bloody “SQL Broker” processes running then dumping everything over ODBC. In order to circumvent the arbitrary and unfair “Users/Connections” licensing restrictions I reversed Progress’ super-seeekrit program code to figure out how their license keys worked and made my own keygen in a weekend, that was fun - I got paid to do that too!). The fact it’s possible to slam out software within a few days while drunk-and/or-stoned (I don’t remember, lol) that undoes Progress’ entire business-model built-around vendor lock-in shows what a house-of-cards they’ve built for themselves.


I would like to say I have a job for you but I am afraid that this kind of decisions is way above my pay grade.

Still... drop me a line in mailbox, maybe? (You can find my contacts in the profile).


VB6 > TLA+


> I do see a lot more intellectual rigor in the older techies than in the ones from my generation

I have a theory on this. An architect in private practice told me a couple of years ago how he was having trouble getting good junior architects, so was resorting to graduates. He felt his small practice was not the best place for a rigourous training. He recounted that in the past there was a giant pool of qualified and experienced staff coming from the councils, railways and public utilities. There was also a similar pool of state supported industry, Rolls Royce, BA etc. These positions took graduates (or even apprentices) and trained them up. These were also jobs for life with opportunity for progression and final salary pensions attached. Most of these have been squeezed out by 30 years of privatisation and The Cuts post 2008.

I personally worked at a company that had grown out of a government research agency some years before. The majority of employees were from that original batch of junior engineers transferring over, even filling roles like HR, Health and Safety, purchasing etc. I don't see how this could happen today.

I feel like in lusting after what Silicon Valley has, we have undone what we had with nothing to replace it.


My prediction is that they will add a licence tooltip in the code completion ui and solve that issue next month.


Github is just the begining, they will crawl any open source code, crawling npm, pypi, cpan, public gitlab...

If your code is open source, they will get it.

That's kinda the point of open source.


Well... "the point of" open source for some of us is to participate in a collaborative commune where we are willing to contribute code to the collective specifically because other people are required to also contribute their code to said collective as part of a shared battle against opaque proprietary systems; so the concept that someone is going to then find it, scrape it, and "launder" what was supposed to be our competitive advantage into their closed source projects by training it into a glorified pattern recognizer and then regurgitating it into their text editor without even as much as attribution without realization that the work we did and contributed to the world was part of a tit-for-tat pact as they want to have their cake and eat it too kind of ruins it for me... like it is bad enough that I have to constantly guilt trip people into not violating my license already: systematizing it via Copilot is making me think that we might need even stricter mechanisms for distribution of code within the commune so as to build a kind of "decentralized trade secret". (That said, other than your wording about "the point of open source", I agree with your doom-y sentiment 100%.)


I’d argue that this new use case is very interesting to open source and how it relates to the various licenses, and not necessarily “the point of open source”.

I can imagine people being OK with their code being used as-is, and/or being modified, but not used completely out of context to train some corporate AI to inject code into commercial code based.


You can't give out something for free without limitations and then complain when someone uses it for something you didn't expect. Well you can complain but no one has to listen.

Githubs use seems very in the spirit of open data and code. Using open source to help others.


We didn't give it away for free without limitations: we carefully drafted licenses to attach to our code to establish legal boundaries on what we considered acceptable; my code, for example, is under GPL, which says "I am OK with you using the products of my labor under the requirement that you will then let me use the product of your labor"... hell: almost every bit of open source code has at minimum "I am OK with you using the products of my labor as long as you are at least willing to give me credit somewhere", a minimal limitation that Copilot doesn't honor (and prevents its users from honoring): Copilot is essentially a giant code launderer designed to strip away licenses from effort that was only contingently given to the community.


Licenses explicitly are the set of permissions and limitations for someone to use the code. Even most permissive open source licenses require that you maintain an attribution with the text of the license in derivative works. Copyleft licenses and in particular the GPL put very strict requirements on what types of licenses are acceptable in any derivative works.

In literally no open source license except "do-whatever-you-want-i-dont-give-a-damn-bye"-type licenses like WTFPL are you giving something away for free without limitations. That's not even close to what open source means, at all. Open source and public domain are not synonymous terms. And in those cases where there are literally any terms in a license, use of Copilot obviously violates the terms of the license, unless the user operating Copilot goes back, finds the code Copilot is stealing from, and makes the proper attribution / otherwise ensures that they are meeting the terms of the license of the stolen code.


Agreed. I am considering relicensing all of my permissively licensed code because of this. The fundamental assumptions I had when releasing that code under a permissive license have been violated.


Under GitHub's legal theory (fair use), nothing you put in that license file can stop them from doing so legally.

If copyright applies, they're already in violation by failing to attribute your MIT contributions and could theoretically be sued for infringement (as they did not abide by the terms of the license).


MIT only requires attribution for "substantial portions". Whatever that is...


When you use a permissive license, it’s best you stop thinking of it as your code. You’ve set it free for everyone, and while you may retain copyright in some abstract sense, it really no longer belongs to you.

That’s a good thing.


While I agree with you--in that the vast majority of people who publish code under permissive licenses actually have an implied set of "moral code" restrictions that they end up surprised people violate even though they specifically allowed such by their choice of license (implying they should have chosen a different, and likely more restrictive, license)--even permissive licenses tend to at least include "you can't take my code without crediting me for taking my code", and so I can appreciate someone being upset about that not happening.


I think hackers being possessive about sequences of uint8s is exactly as silly as claiming the number 2 as your own personal property.

Copyright is like an ABI: we made it up for convenience. It's just a construction, it isn't some inherently real thing.


Water pollution too.


Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: