Archive for Controversy

21 results.

Slate posts a very bad proposal for Twitter users and shareholders both.

Oct22
on 2016/10/22 at 7:02 pm
Posted In: Business, Controversy, Google, Identity, Privacy, Real Life, Twitter, Virtual Worlds

There’s a new article on Slate, by Will Oremus, entitled “Verify, Then Trust”, subtitled “How Twitter could finally solve its abuse problem“. I think it completely misses its mark.

First, it references another Slate article by David Auerbach that … actually doesn’t suggest anything with identities or verification at all.  Oremus suggests the Twitter problem could be solved one of two ways: the Auerbach way (which is a dramatic change, but a practical one focusing on the actual real problem), or the Oremus way (which naively equates anonymity with trolling, sticks its head in the sand, and ignores the root problems).

Auerbach points out (I believe quite validly) that part of Twitter’s problem is the need to try to show simple growth to shareholders, since going public with the company a few years ago.  And this actually varies greatly from the goal of providing the best service, and a service of value to the user.

It has been argued that revenue for social sites comes from knowing who the user is.  In “real life” (RL).  I disagree.  Money comes from providing the identity service, not for a specific identity, not from knowing real-life information. It comes from seeing that user Aardvark Alphabet was shopping on Amazon for cordless drills or whatever, then when Aardvark Alphabet later does a search for building supplies, or some other social activity such as posting a picture of something Aardvark has built, it can pull out ads for companies selling cordless drills and insert them into his search results or any other ads Aardvark is being presented with.

It doesn’t matter who he is, or who he has authenticated as, on other sites.  The idea is to be the one to provide that service that allowed him to log in as Aardvark Alphabet in the first place, or any identity.  But just to be his identity provider… i.e. to also be his intelligent, targeted ad provider.  And that is just for the ad revenue part of this.

But more importantly, there are many many reasons for using an alternative online persona, and any policy that doesn’t recognize that is alienating a percentage of its users.  Here’s the EFF summary explaining it: https://www.eff.org/issues/anonymity If you think online pseudonyms are a bad thing, read that first, it’s not long.

And here’s the Google policy that shows good insight on this as well — even though they too battled against their own policy with Google+ identities: https://publicpolicy.googleblog.com/2011/02/freedom-to-be-who-you-want-to-be.html

Now, like Google, the author of this article too was confused into falsely thinking anonymity had some correlation with online trolls; it does correlate, but inversely!  Trolls are often worse under real-name policies. See this article discussing one such study: https://www.techdirt.com/articles/20160729/23305535110/study-trolls-are-even-worse-when-using-real-names.shtml

This paragraph in particular is almost frightening in its self-contradition:

“From Twitter’s perspective, my plan would probably run afoul of its noble, if arguably misguided, hard-line commitment to anonymity and freewheeling speech—all without fully solving the abuse problem. It’s true that there’s great public value in a platform that allows almost anyone to be heard, even if others would like them silenced. Twitter’s importance to political dissidents, for example, is underscored every time an autocratic regime tries to censor it or shut it down.”

I’m flabbergasted at that first sentence. I’ll take a closer look at it below.  Andrea (Twitter user ‏@puellavulnerata, who is a core developer for the Tor anonymity software) wrote:

“Please do explain, @WillOremus, what constitutes a ‘legitimate’ reason for pseudonymity and why you failed to notice that it isn’t pseudonymity anymore if you have to leak identifying details to prove you have one.”

If the content is not self-governed in some way by the users of the service, then Twitter has failed in its primary mission.  There cannot be any kind of central authority choosing sides. As Andrea put it:

“Twitter has misconceptualized the problem as ‘good people’ vs. ‘trolls’ when it’s more like intercommunity conflict and so they’ve built tools which embed assumptions that everyone wants the same stuff filtered far more than is actually true. I don’t care about investor perception, I care about having a place to conduct my online social life that won’t get destroyed out from under me by some idiot fixation on legal identities and universalized norms of conduct. Hence, in the long run, we must kill centralized platforms like Twitter and build a replacement not dependent on investor whims.”

Whether we need to a replacement for Twitter or not, I’m not sure at this point. I do know that any real-name policy will kill Twitter.  It runs contrary to its very fiber and I agree with Oremus above stating his own plan would effectively be a horrible idea: “From Twitter’s perspective, my plan would probably run afoul of its noble, if arguably misguided, hard-line commitment to anonymity and freewheeling speech—all without fully solving the abuse problem.”

To paraphase that: “My plan would ruin the very reason for Twitter’s existence, and wouldn’t actually solve the problem.” Ohhhh-kay. The cost-benefit trade-off is off the scale.  So if they do that for the investors, good luck with that share price a year or two later.

I do agree with Oremus’ final two paragraphs. Twitter is different than Facebook and Google+, and needs to stay that way:

“Unfortunately, Twitter is not a public-benefit corporation. Since it decided to go public in 2013—a mistake, I believe, in retrospect—the company must answer to its shareholders, and they’ve made their top priority clear: growth. And not slow, steady growth, but rapid growth on a massive scale. They want Twitter to be more like Facebook.

“I’ve argued for years that Twitter is fundamentally different from Facebook, and we should all root for it to stay that way. Yet, for all its shortcomings as a venue for discussion, Facebook’s platform is far less conducive than Twitter’s to public abuse from unaccountable trolls. At this juncture, Twitter simply has to find a way to become a little less hostile. If it can do that, at the very least the company will become palatable to corporate suitors while better serving the majority of its users. At the same time, its data will become more valuable to advertisers. And beyond that, who knows: It might even start growing again.”

So why would a company want to be the identity service for a smaller portion of the population?  Why can’t they learn from these same mistakes made by Google (e.g. Google+ and YouTube both) and others, and just let users choose how much personal information to reveal?  Because they (blindly and falsely) think anonymity means trolls.  So they go ahead and exclude a percentage of the world population that needs that social connectivity the most.  And in addition to being a social crime against a subset of the population, it’s also not in their shareholders’ best interests.   #fail

The good news is that this is just a proposal from someone writing for Slate. It’s not a good one, and it’s not a proposal from Twitter, so there’s no reason to believe Twitter would pay any attention to it. Let’s hope they stay the course on anonymity and the use of online pseudonym.

Comments Off on Slate posts a very bad proposal for Twitter users and shareholders both.

Some Overlooked *Good* News on FCC Net Neutrality Ruling

Jan15
on 2014/01/15 at 12:01 pm
Posted In: Business, Controversy, Politics, Technology, Trends

There has been quite a bit of doom and gloom regarding yesterday’s decision by the United States Court of Appeals in Verizon v. FCC, however I feel there are two significant aspects of that ruling that have actually ensured a future of net neutrality.  Yes, on the surface it sucketh greatly, but the more I think about the ruling, the more I believe it could hardly have gone any better for the FCC.  It is pretty much the worst possible scenario for Verizon and the other Internet providers, other than a total loss on every aspect.

There are two very critical positive aspects to the ruling:

  • disclosure
  • authority

Any losses to net neutrality are either short-lived due to the second aspect, and significantly weakened by the first.

 

Disclosure

Yesterday the court ruled in favor of the FCC when it comes to providers being required to disclose any non-neutral traffic management activities:

The ruling did, however, preserve the FCC’s current power to require Verizon and other broadband obligations to disclose their activities — in other words, to reveal how they are managing traffic: “Verizon does argue that the disclosure rules are not severable, insisting that if the anti-discrimination and anti-blocking rules fall so too must the disclosure requirements. We disagree […] we are satisfied that the Commission would have adopted the disclosure rules absent the rules we now vacate, which, we agree, operate independently.” (see the full gigaom.com article)

This is actually very bad news for Internet providers in that it means those with complaints against providers may no longer have to figure out, and prove, what the providers are doing.  And if the providers do more than they disclose and that is discovered (and eventually would be), then they would be found in violation of this ruling.  And the second point above makes that situation very interesting.

 

Authority

Yesterday’s ruling also made it clear that the FCC does have the authority to regulate broadband (as well as its understood authority over carriers):

The biggest piece of good news is this: On the issue the case was supposedly about, the one with the greatest long-term significance, the FCC won. Verizon’s core argument was that the FCC doesn’t have legal authority to regulate broadband. The court’s answer was unequivocal: It does. The FCC made a sufficiently strong case, the judges said, that it could do so to promote investment in advanced communications services under Section 706 of the Communications Act. And, they continued, the FCC was sufficiently convincing that net neutrality served that goal. (see the full theatlantic.com article)

This is both critical to the long-term, as the technology changes, but also in combination with the disclosure ruling above, it means that the FCC has the authority to act when violations of the disclosure occur.

 

The “Net” Result

Overall, I believe this leaves Internet providers are placed in a very awkward position: they may now be able to favor their own websites and media sources over third-party ones, but they will have to reveal that they are doing that. And again for what may be their larger goal, they can now charge extra for fast, reliable service, but they have to reveal that.  And the FCC still has authority over these Internet providers, and retains authority over future Internet services.

The first provider to announce different levels of service, or preferential website treatment, will likely suffer a significant PR backlash, and customers (and the FCC) may be silently thinking “Go ahead, punk. Make my day.”

Verizon’s official statement after the ruling may have been worded to try to deal with the potentially huge negative public relations reaction:

Verizon has been and remains committed to the open Internet which provides consumers with competitive choices and unblocked access to lawful websites and content when, where, and how they want. This will not change in light of the court’s decision.

Verizon seems to be saying that the court has ruled that the FCC cannot block opponents of net neutrality, however Verizon will voluntarily comply.  But a skeptic’s view of the statement would highlight the key word “unblocked access”, which is very different from “neutral delivery of”, including performance differences.

Because of that wording, I remain one of those skeptics.  I will accept the good news in the ruling, but we must all ensure that the disclosure aspect of the ruling becomes the primary tool in the fight to preserve net neutrality.

Comments Off on Some Overlooked *Good* News on FCC Net Neutrality Ruling

KFNX Responds – Barbara Espinosa Loses Her Show

Jun22
on 2012/06/22 at 12:41 am
Posted In: Controversy, Politics

You’ve probably heard the story of Barbara Espinosa, who ran a radio talk show with the former Arizona GOP chairman Randy Pullen called “Hair on Fire”, who recently gained international attention when she called U.S. President Barack Obama a “monkey”.  Repeatedly.  She then followed that with “I voted for the white guy.”

Arizona GOP Radio Host Calls Obama First Monkey President

Here is a longer MP3 recording of this mess, which includes the rejection of these comments by another caller, followed by both callers hanging up:

Hair-on-Fire-News-Talk-Radio-030112.mp3

Fortunately, at 1:58 in this recording, the other caller completely rejects Espinosa’s comments.  And that ends that.

I contacted the radio station where the broadcast was believed to have taken place.  They report that the host in question “does not host a show anymore” at that station, and that the show in question may have happened elsewhere since she has not worked there in a month.  However, based on the filename of the recording, which has the suffix “030112.mp3” and would appear to be a date, it would seem that this is a recording of a show that aired months ago, when it would have most likely aired on KFNX.

In their response, the station has clearly moved to distance itself from Espinosa and the controversy:

KFNX no longer has a relationship with Ms. Espinosa, and again certainly does not support her comments. If those comments were made on KFNX, we would have terminated our relationship with her. We certainly apologize for our former relationship with Ms. Espinosa and are deeply sorry she said offensives things.

Here is the full text from KFNX:

Statement from KFNX 1100 RE: Barbara Espinosa (6/18/2012)

KFNX received your email. Barbara Espinosa does not host a show anymore on KFNX 1100, so the information is dated. She has not aired a show at KFNX for nearly a month. She currently hosts an internet show on Blogtalk radio to the best of our knowledge.

KFNX management, staff and sponsors do not endorse or agree with her viewpoints. Ms. Espinoza (like all KFNX Hosts) does have the First Amendment Right to say what she believes, but KFNX Host Contracts includes a clause which prohibits on-air slander of people.

KFNX no longer has a relationship with Ms. Espinosa, and again certainly does not support her comments. If those comments were made on KFNX, we would have terminated our relationship with her. We certainly apologize for our former relationship with Ms. Espinosa and are deeply sorry she said offensives things.

We do not know when the comments were made, but it is possible they were made in the past while airing on KFNX, or maybe could have been said on her current internet show on Blogtalk. We also do not know who posted the YouTube video. It is an edited video (and not dated), so it is unclear what context the improper comments were made.

If you want more information, we suggest you contact Ms. Espinosa (or Blogtalk) directly regarding this.

Again we apologize for any inconvenience.

Sincerely,

KFNX Management

 

2 Comments

Understanding Scripted Region Crossings

Feb29
on 2012/02/29 at 1:50 am
Posted In: Controversy, Grids, InWorldz, Linden Lab, Second Life, Technology, Trends, Virtual Worlds

Over on Inara Pey’s blog posting on region crossings by vehicles, Pussycat Catnip added a comment that asked the question:

Are there really Open Sims that -lack- this ability?  I’d just assumed this was as standard as the ability to rez-in your own avatar… (ie: logging in).

I started to post a reply there, but after seeing the length, I did not want to hijack that blog posting in any way. So here is my answer here:

No, to the best of my knowledge, all OpenSim grids have lacked this ability until now, unless you include InWorldz.  It’s a very significant thing, and was in also a big accomplishment by Linden Lab when they provided this for MONO scripts in Second Life.  But it’s not really anything to do with physics, or vehicles.  It’s about transitioning a running script (in anything) from one region to another.  The explanation is a bit long; my apologies.

The Script Continuation Problem

When SL went with the .NET/MONO runtime environment, they forego the ability to control scripts very closely, since the runtime environment was developed by a third party.  But during a region hand-off, they need complete control.  They need to be able to stop scripts running in one virtual runtime, transmit them to another region, and load them in an object on that region, and not restart the scripts, but rather continue them from where they left off. So it’s just not a matter of re-rezzing a new copy of the same thing on another region; it must restore a copy of that object with the scripts running in the same context as they were when the object hit the region border.  All the active data, the current state of the script’s execution contents, must be restore and continued from where it left off.  For example, before this, in OpenSim, scripts were restarted (from their beginning, losing their current context) after a TP or region crossing.  InWorldz achieved this continuation of scripts with the introduction of the Phlox script engine last summer, but it was not possible in OpenSim until now.

Linden Lab’s Solution

When Linden Lab implemented this before any of the alternative grids, there was no ability to get this execution context from the third-party runtime environment (MONO). So Linden Lab developers effectively had to become MONO developers, and provide significant hacks, er, I mean extensions, to the MONO project.  Which then effectively meant they were running their own variant of MONO.  (I’m not aware of how extensive or localized the changes were from the standard tree.)  It was a lot of hard work, but they eventually provided the hooks in MONO that they needed to pause running code (MONO scripts compile to native code), and to serialize the data into a stream that could them be fed to the next region, deserialized, and reapplied to a copy of the object on the other region.  That was a lot of work and I’m sure when they started, they probably weren’t really sure to what degree their success would be.

InWorldz’ Solution

When InWorldz chose to attack this problem about a year ago, they chose a completely different path. They chose to create their own Phlox Script Engine runtime environment as a virtual machine, providing whatever hooks were needed as an inherent part of the design of that virtual machine.  Then compile LSL (or any language desired) into the intermediate p-code that their own virtual machine understood.  Not only does this keep control of the runtime environment for scripts within the InWorldz development project, but it allows much easier extensions in the future, much MUCH better processor consumption and memory management, and complete control of scripts.

One of the planned side-effects of this Phlox design is direct control of hand-offs between regions.  These can be done *so* efficiently that some naysayers actually complain that videos of crossings (here and here) must have been doctored, or faked in some way.  It’s that good.  Vehicles are just one example of script crossings. Physics really doesn’t play into this much.  If you can walk across regions, you’ve performed complex crossings of physical objects.  The tricky part is having the scripts continue where they left off, uninterrupted, and this includes vehicle scripts (and a lot more).

Avination and OpenSim Milestone

This is why I give credit to Avination for their recent success in the major work item of script state persistence across region crossings.  The vehicle part of it isn’t really the big deal here.  Having true continuation of active scripts on the other side is a Big Deal.  This is also a key part of why I found the “first” claim by Avination to be so outrageous; InWorldz has had vehicle crossings since the Phlox runtime environment came online last summer.  Even in terms of physical vehicle crossings, InWorldz has had ODE physics (same as other OpenSim grids) since before InWorldz was founded.  However due to its ability to cause region crashes, it has been disabled for about a year (out of InWorldz’ 3 year history).  If InWorldz simply turned physical objects back on, physical vehicle crossings would have been possible since the introduction of Phlox; the hard part, and the part Avination just completed, was the persistence of the scripts across the crossings, and that was fully functional and available grid-wide in InWorldz last summer.

However, since the Linden Lab proprietary implementation, another third party developer has provided an implementation of continuations in MONO, which has been available since MONO 2.6.  This provides the context save/restore needed, and Avination has successfully applied that to the OpenSim runtime. That’s great news for MONO-based script engines in OpenSim.

Thinking Ahead

In the long run, I see moving away from MONO as the best (and perhaps the only) way to tame processor and memory use and provide complete control of the runtime environment. Phlox will provide that total control, which means more features. Things like much easier support for new languages, an LSL debugger that can be built-in to the viewers to allow single-stepping through LSL code and examination of variables at each stage, easier control over scripts CPU and memory usage, and many other secondary benefits, such as no need to try to limit script cost artificially and blame script authors for hurting sim performance.

But the bottom line here is that continuing a script in a new region, running on a completely different machine (IP address) is a Big Deal.  Physical objects, a vehicles, not so much, but persisting those scripts, hell yes.  It’s a major accomplishment, and now OpenSim has it too.

14 Comments