Hacker Newsnew | comments | show | ask | jobs | submit | josephlord's comments login

Be aware that you can't family share in-app purchases (even persistant upgrade type purchases).

-----


All banks create money out of thin air when they make loans not just central banks.

And your arithmetic doesn't quite work either at the very least if you assume that money actually gets respect with a certain speed.

http://www.forbes.com/sites/stevekeen/2015/03/30/the-princip...

-----


for "respect" read "respent"

-----


Atheism is not a doctrine any more than not believing in Santa Claus or the tooth fairy is a doctrine. It is possible that there are doctrines within atheism but atheism itself is not one.

I'm not bothered about the term free thinker and wouldn't necessarily have linked it with atheism.

-----


I'm aware atheism isn't a doctrine - I said within atheism. Albeit said with an annoying double-word typo. :/

Anyway, if you don't think 'freethinker' is linked to atheism, throw the word into google and skim the results.

-----


See recent Lenovo issue. It isn't a user problem but an ecosystem problem starting with the OEMs and the general discovery and distribution of software.

-----


She is clearly receiving harassment and is prepared to publish it, why the hell wouldn't they bring her into it? And if you think she gets too much exposure why did you mention her?

I'm only aware of her because of the harassment her and others have received. The people trying to silence her with abuse are amplifying their opponents and discrediting themselves and proving the need for change in these areas. There are clearly some people going out of their way to try to bully vocal women into silence and I'm glad it isn't working.

-----


The "relentless" part is critical for whether insults (or even just continued asking of questions or repeating points) are harassment. Especially if the insult isn't profane - e.g. "idiot". To what extent is it OK to tell someone they are being an "idiot" or "are an idiot" when they say something idiotic. Some people have even taken to being insulted by being called "sexist" or "racist".

The difficulty comes when someone is relentlessly harassed by a large group such that each individual's insults on their own wouldn't be harassment. The recipient is being harassed but I'm not sure what the correct response is in this case. Also it is difficult to see the size of the group and who is actually involved if there are many throwaway anonymous accounts. There is also a hard to make distinction between a large group responding to something and calling it out following it being retweeted (which might briefly feel like harrassment) and a group organising and deliberately trolling people or jumping into every conversation they have.

These are tricky questions and I don't have answers to everything. Threats of violence while horrible are at least a simple and clear cut case that can be responded to (ideally with at least a police visit/warning rather than just a ban from the service). That does not make them the only sort of real harassment but it is in a way easy to deal with.

-----


The "relentless" part is critical for whether insults (or even just continued asking of questions or repeating points) are harassment. Especially if the insult isn't profane - e.g. "idiot". To what extent is it OK to tell someone they are being an "idiot" or "are an idiot" when they say something idiotic. Some people have even taken to being insulted by being called "sexist" or "racist".

This is another sort of claim that pops up a lot in freeze peach[1] debates, where we pretend that reasonable people do not exist, and treat all claims as somehow equal and thus intractable. This is complete nonsense, and something the sane society largely doesn't have as much of a problem with as the internet does, at least as argued by those still desperate to defend the right to an audience for invective.

[1]http://www.patheos.com/blogs/daylightatheism/2013/05/free-sp...

-----


I do agree with you that it can be picked apart but it is very hard to do with simple rules.

Society does have problems with it in many areas but most noticeably to me around the intersections of race, religion and politics. Can criticising the Israeli government be anti-semitic? (It probably can be but the accusation is levelled as a defence against normal criticism). At what point does criticism of Islam or particular practices become problematic, Islamaphobic or racist? There isn't a clear line that everyone can agree on.

I think we probably agree on the answers we would like for these things and especially our views about "those still desperate to defend the right to an audience for invective" but it isn't always easy to set rules and operate systems to keep it working.

-----


I think anonymous accounts should be expensive (in terms of time) to set up. Anonymity and pseudonyms are important for those who need to whistleblow or are from oppressed groups so the possibility absolutely needs to be preserved. But what isn't needed is cheap throwaway accounts that can be used for abuse and then discarded as soon as they are banned.

If to sign up anonymously you had to do something, a quiz or play through a game that took about 30 minutes then that would reduce the rate of account creation for abuse. If you were prepared to give a real phone number (and use it for verification) then you could bypass the task and get an instant account but obviously any ban would apply to the phone number not just that particular account.

-----


There are plenty of people who are fine being vile human beings on their normal accounts under their real names.

http://www.salon.com/2014/12/10/whats_up_with_the_rag_head_w...

> “What terrorist group does this guy belong too?”

> “this is horrible, Facebook should be promoted by a different person, i have lost a lot of respect for Facebook because of there choice to use this person.”

> “Can I borrow your towel on your head I ran out of toilet paper please sir?”

> “What’s up with the rag head?”

> “Get that Isis terrorist off of your ad”

> “Get this camel jockey off my news feed pronto!”

-----


And they can be banned with a real cost to the owner. The decision about whether or not particular things cross the ban line is separate from the design of a system that imposes a real impact when the ban occurs.

Also in the context of harassment they can be blocked by the person being harassed

-----


Teamspeak v3† solves this in an interesting way: the program supports creation of anonymous (or rather, pseudonymous) identities easily, each associated with an unique ID which takes adjustable amount of time to compute on users' computers.

The adjustable parameter, called "Security Level", is expressed as a small integer.

To protect from flood of new (spam) identities, channel owners / operators can select minimum difficulty allowed on the channel. The higher the Security Level required, the more time it takes to create such identity. Values around 24 represent a dozen seconds worth of computation, while around 28 is a few minutes.‡

This easily discourages casual spammers, while legit users have no problem with spending a few dozen seconds, or a few minutes to create an identity.

† VoIP team chat, used by gamers.

‡ more information: http://forum.teamspeak.com/showthread.php/47502-Security-lev...

-----


Exactly this, and what's so mind boggling is that this is not a new problem by any means. I'm sure that among twitter's worries is how to effectively scale an abuse prevention operation, but there are some pretty easy automatic filters that would go a long way. A few that come to mind (provided accounts are tied to phone numbers):

1) A block that prevents the harasser from even @mentioning the harassee in any way.

2) Automatic disabling of accounts that are blocked by more than two users whom the harasser explicitly @mentioned

3) A "dog house" mode which automatically disables distribution of a user's tweets who have been reported as harassment without explicitly telling them, leaving them to harass nothingness until they get bored and leave.

-----


Blocking is an issue, since that can be weaponized way too easily: "I don't like your non-harrassing opinion, so I'll get two buddies to engage with you, then block you once you respond."

The "dog house" could even work for cliques: If a certain number or percentage of your peers (as determined by follow relationships and metions) block a user (for themselves), the platform could make them disappear for you as well (unless you engage with them first).

That should be good enough to keep adverse groups separated while avoiding the issue of "I'm not abusing them, I'm calling them out" that's already mentioned elsewhere.

-----


1) kind of already happens, they can mention but it's not linked, which is laughably enough to prevent a significant degree of mobbing by their own followers.

2) Someone in the atheist-skeptic community made a tool called the blockbot that ostensibly was to automatically block people from a list that you never want to talk to in the first place because the blockbot operator already identified them as trolls. Unfortunately the main reason was that the developer figured out that mass blocks caused an automated response mechanism to suspend that users. It was really an offensive, not defensive mechanism. This has been somewhat changed, but the reason I bring it up is that this has been tried by Twitter and unscrupulous people used it as an attack tool.

3) is similar to 2) but I suspect they are already doing this in some cases.

-----


Number 2 could probably be abused. Get several people to respond to somebody and when they reply block them.

Number 3 needs to be applied carefully as wrongfully banned people (it does happen) will be harmed.

-----


(replying to both since they're effectively the same comment)

Abuse of such things is definitely a challenge, and there is finesse on choosing the right thresholds. I doubt twitter can fully escape manual moderation and as such one of the goals of a system like this would simply be to reduce the number of situations that require human intervention. Undoubtedly, there will be false positives, but abuse of the abuse system will likely be rarer than abuse as a whole and if a human moderates and finds such meta-abuse it would be grounds for the strictest means of blocking twitter can provide, thereby hopefully incentivizing against it.

-----


What if a user could toggle a setting through which users of a certain type couldn't @reply them at all? e.g., accounts created within the last month, or accounts with fewer than 50 tweets, etc.

-----


I don't think anonymity is the real issue, although it makes trolling easier. Services such as Facebook and Youtube require real names, yet they still have lots of people trolling each other. I think it is rather the feeling of being able to get away with it which causes people to have an illusion of anonymity, whether or not there name/phone number is actually associated with their account.

I'm curious if it isn't feasible to do traffic analysis and nip trollers in the bud when it turns out that one or more twitterers are sending a large number of tweets to a single person without an actual conversation taking place.

-----


> I don't think anonymity is the real issue, although it makes trolling easier.

Anonymity isn't the real issue (though it's a compounding factor), the issue is how easy throwaway accounts are to create: for victims of abuse, it's not possible to "manage" abuse by blocking abuser accounts because within seconds they can be on a new yet-to-be-blocked account. In fact, I'd expect most serial abusers don't even bother waiting, they create an account, throw a salvo of abuse, create a new account, new salvo, and the account creation can be trivially automated.

If accounts are either cheap and unique (based on "real-world" identity) or relatively expensive (time-wise) but anonymous, the ability to block people becomes far more useful for the serially abused.

-----


Like the governments that promoting security and safety at the expense of privacy and freedom, Twitter has a too bad a record of dealing with internal abuse to be trusted. For example, there were stories about them stealing usernames from people (because a celebrity showed up). Similarly, any kind of process they set up for dealing with external abuse can be abused; people will be flagging accounts they disagree with, hoping they would be banned, and other people would DDOS the system by flagging too many accounts, making it impossible for Twitter to keep up with it.

The solution is simple: set up a new service, see if users use it.

-----


Definitely worth considering but any significant impedance to joining the network would have a financial cost for Twitter in terms of adoption, and would never get up.

-----


Swift passes the test in my view provided you can cope with using the word struct instead of object. (Objects based on classes are reference types in Swift but structs, enums and the default collections are value types).

You can write pretty functional code if you want but you also have the option to write more mutable/side effecty/stateful object code if you want.

-----


So does Rust, I think. 2014 was a good year for programming languages.

-----


> 2014 was a good year for programming languages

Yeah... another batch of innovative, interesting languages died because "it's not practical", another pile of language features didn't manage to get implemented because "but look at syntax, it's ugly", another bunch of languages took some decades old features, dumbed them down as much as possible and then some more (because "it would be to hard to use otherwise") and marketed them as ground-breaking improvements to the art of programming.

Very good year for PLs, indeed.

-----


Have you got any concrete examples, or are you just going to complain?

The entire field of programming language design is more complex than you seem to be admitting.

-----


Nice syntax and easily explained features are extremely important parts of programming language design. A really fancy new feature whose implementation compromises these qualities is usually a step backwards for the language as a whole. Look at C++.

-----


Last I checked you can't quite get away with using struct only in Swift because you need objects for indirection in recursive structs.

But other than that I really, really hope that struct-only coding takes over in Swift given time... perhaps lots of time.

-----


Fair point the need to Box things is a bit of a nuisance. Also staying pure and immutable gets tricky as soon as you hit the Cocoa APIs but it is definitely a step in the correct direction and I still like it.

-----


It might well be that it was a target price but there is actually a fair chance that the company may have achieved it by tightening their margins rather than expanding them.

Source working for a TV manufacturer and knowing how desperate the sales teams were to get prices to hit certain magic "price points".

I'm not saying that it always works this way though.

-----


Especially true with things like TVs, where sale prices are often loss leaders with the stores planning to make it up by selling lots of overpriced cables and other accessories to the buyers.

-----


> Land Value tax makes the rich pay a lot but hits the poorest really hard since the threshold to owning land (and creating wealth) is substantially raised.

That seems a bizarre statement. Firstly I don't see any connection between owning land and creating wealth. Secondly taxing land value is a deterrent to land ownership and should reduce the desirability of owning land and reduce the price of land. This will make it more cheaply available to productive use and reduce the incentive to hoard land unproductively. Finally the poorest cannot afford to own land anyway so are not affected by the land taxes.

Have I missed something?

BTW I largely agree about cheating VAT especially in a small country with neighbouring countries that can be cheaply visited.

-----


What I'm trying to say is that social mobility will be severely stunted.

We can agree that the reason for owning land is to make money, yes? Planting crops would take a couple of months to return the initial investment. If I planted a forest it could take years.

If this investment possibility is removed from everyone without (massive amounts of, if the tax is high enough) capital it's going to create a further wealth gap for no apparent reason.

> Finally the poorest cannot afford to own land anyway so are not affected by the land taxes.

Yes, this is true for the poorest, bad phrasing on my part, but it is going to make the barrier of entry higher.

Also: land produces goods that everyone need the same amount of (food, wood, wool), this would increase prices in a way that wouldn't really bother the capital heavy but would seriously hurt the already poor.

-----


Land can make money purely by rising value over time so people can speculate by owning land. There may be little incentive to actually use the land.

The ROI on a forest takes years whether or not the land value is taxed. You also said earlier that it was a tax on land value so presumably much lower for forestry land than urban land.

The primary impact of a land value tax is likely to be on the price of land itself because land will be less profitable. Introduction or increase will hurt those who own land (and its abolition would benefit land owners).

If we assume that the cost of access to land (rent) is set by the free market then we can assume that demand is essentially unchanged and the supply of land would if anything be increased (if some speculators sell or start to rent out their land) so access to land for the less wealthy should be cheaper.

I think your assumption that ownership is the way that less wealthy access land is fundamentally wrong. I think that this will also apply to the effect on prices of things made on the land on which I expect little impact (although there may be side effects on the rental market that make this more complicated).

-----

More

Guidelines | FAQ | Support | API | Security | Lists | Bookmarklet | DMCA | Apply to YC | Contact

Search: