Cover photo
David Gerard
Works at Spartacus
Attends Spartacus
Lives in Spartacus
2,143 followers|905,922 views


David Gerard

Shared publicly  - 
I have a Tumblr now. Had it for a week or two, post about 10x a day to it.
kill all the humans I am here solely to follow my purple-haired Tumblr feminist SJW spouse’s slashy...
Edward Morbius's profile photoGod Emperor Lionel Lauer's profile photoDavid Gerard's profile photo
Lionel: how annoying. Note added to post.
Add a comment...

David Gerard

Shared publicly  - 
So I just looked at download numbers, and it turns out AOO has knowingly distributed at least 8 million vulnerable copies of AOO since April 27. Rather than remove one file to fix it, they ... do nothing.

You have computer-illiterate friends and family who probably have AOO. Please help them out of the hole AOO has left them in.
Matthias Seidel's profile photoDavid Gerard's profile photo
You came to my G+ and you're asking me what my problem is?
Add a comment...

David Gerard

Shared publicly  - 
I'd like to echo:

"I'd also like to add, since the point seems to get lost in discussions:
⚫ I use G+ heavily, and have since early Beta days, June, 2011.
⚫ I find it useful."

"Quantitative numbers for G+ performance and activity have greatly lagged other networks. Over the past year and some, Google have progressively de-emphasized the social network in marketing, discussion, and systems.
Playing this as some cryptic sign of strength is profoundly unconvincing."

Accuracy is important.
Death of Google+? Film at 11

The question of G+'s death is being bandied about again, and it's one that deserves asking. I've been behind a fair bit of the discussion myself[1].

I'd also like to add, since the point seems to get lost in discussions:

⚫ I use G+ heavily, and have since early Beta days, June, 2011.

⚫ I find it useful.

⚫ My concerns with G+ and Google fall into three general categories: 1) a failure to live up to expectations for discussion, 2) grave misgivings over privacy and trust, and 3) numerous UI/UX failures. All indicate places Google could improve. And I really wish it would, though four years in, I've largely lost any hope it will.

⚫ I also use several other sites, some heavily, including Reddit, Hacker News, and Ello. Though not Facebook -- my trust issues with it are far greater than with Google.

And specifically this: with a highly curated set of profiles followed, and a no-regrets policy to blocking fuckwits[2], the G+ experience can be pretty good. Not excellent, mind. But reasonably compelling.

That said...

Quantitative numbers for G+ performance and activity have greatly lagged other networks. Over the past year and some, Google have progressively de-emphasized the social network in marketing, discussion, and systems.

Playing this as some cryptic sign of strength is profoundly unconvincing.

Google have, from the very beginning, been at best coy, and frequently manifestly deceptive in activity, engagement, participation, and referral activity concerning G+. The rationale for my own investigations has been to demonstrate that the numbers are fairly apparent if you look for them. Which means they're simply not fooling anyone who matters. Possibly their own fanatical users, and employees. But not marketers, advertisers, investors, business partners, or competitors.

There's the frequent claim that G+ has some hidden strengths. Sadly, if the issue is public discussion, then you do far better to advertise your strengths, loudly. No, Communities aren't some vast wealth of discussion (see previous studies). No, there's not some vast depth of private discussion (though yes, I've found limited private discussion useful). Widespread collaborative discussion tends to strongly favor open rather than closed networks, though not entirely without standards. Noise and referral mechanisms are conspicuously opaque, and always have been.[3]

There's the claim that G+ discussion is somehow "better" or "higher quality" than Facebook. I've been taking a stab at a direct measure of "tracking the conversation" on various terms across multiple sites.[4] I've expanded this to a more general "Rating the Socials" study, which compares Google results for the first 36 of Foreign Policy magazine's "FP Top 100 Global Thinkers" on multiple domains[5], vs. a publicly notable woman largely famous for being famous:

Rating the Socials: Foreign Policy Top 100 36 Global Thinkers vs. KK

(The accompanying image highlights results from this study.)

Tested are Twitter, Facebook, Wordpress, Google+, LiveJournal, Quora, Metafilter, Medium, and Ello.

The first standout result is that, if you're interested in discussion or focused intelligent content, you don't want either G+ or Facebook, but Reddit (discussion) and good old fashioned blogs (content). I'm shocked, shocked to learn that interest-oriented discussion sites favour discussion, and that long-form content sites favour long-form content.

There's also what seems to be an opportunity for a system which works far better to tie together disparate blogs with intelligent commentators across the Internet. Oh, say, something like Google Fucking Reader. But no, that had to get killed. I've suggested to Ello's team that they might want to consider the space -- the social graph there is tiny, but the site has numerous features which might make the concept attractive.

Another element to come out of Rating the Socials is the FP:KK ratio. That's a comparison of the ratio of FP-36 posts per 1,000 pages (see the report below for why), and the KK posts per 1,000 pages. A higher FP:KK ratio means more FP Global Top 36 results relative to KK.

An estimate of total pages per site is obtained using a search for "this" (a common, unbiased, English word). G+, Reddit, and Wordpress all have roughly 1/15 the total page count of Facebook by this metric.

The surprise for me is that G+ is far more ass-obsessed than Facebook: a 3.45 KK/1,000pg ratio vs. 0.58 for Team Zuckerberg. This shows in the FP:KK ratio, where Facebook does 5.38x better than G+ in terms of discussing the Global Thinkers. Google's FP:KK ratio is the lowest of the lot.

So much for Highbrow....

The best FP:KK ratio by the way: Metafilter, at 32.75.

The full list, ranked highest to lowest:

⚫ Metafilter: 32.75
⚫ Quora: 6.80
⚫ Reddit: 4.52
⚫ Wordpress: 2.26
⚫ Facebook: 2.10
⚫ Ello: 1.06 (estimated)[6]
⚫ Twitter: 0.94
⚫ Medium: 0.73
⚫ LiveJournal: 0.55
⚫ Google+: 0.39

Of course, raw numbers aren't all there is. There's the question of how accessible content is, and how easy it is to find (though as noted, G+'s Noise, Notifications, and Search weaknesses are substantial). There's the question of how long any given discussion is live. Here G+ has an edge: threads are never locked due to time, though there's a reply-limit maximum of 500 comments. There's the question of domain-expert depth. G+ suffers here in that with what seems to be perhaps 10% the participation of FB, domain experts on any given topic may not be present. Some are (notably in the Linux community), but this tends to be the exception.

There's also the question of presentation capabilities. The entire reason I posted my findings to Ello, and not G+, is because of the ability to include tables, multiple images (editable after publishing), sections, lists, inline links, and the like. If I'm going to discuss complex topics, Ello makes a far better blogging platform than G+.[7] Its superior capacity for images and embeds makes it more attractive than Reddit (inline images supported only under the RES browser extensions) or many blogs (independent image hosting required).

What would make G+ better?

Making it a chosen option rather than a forced one would help a lot.

Better publishing tools. Markdown (I requested that of Yonatan years ago). Or better yet: user's choice of markup language, appropriately sanitized. Multi-image support. Embeds. Tables.


This isn't fucking rocket science. Blogs have had this forever, it's called "tags". A stock set of content tags, and the ability to subscribe to these, would be Really Fucking Slick.[8]

Oh, and classification by language (foreign posters may be of interest, their foreign-language posts generally less so, despite G+'s quite good inline translation). By media type (images, animations, video, music). Location, where appropriate.

RSS/Atom feeds. Incoming and outgoing. G+-as-feedreader could be a thing. With Sparks (an early feature, long since ditched), it had elements of this.

Better Search. The comprehensiveness and speed are good. The specificity and controls are fucking idiotic. The fact that Google launched a service without search capabilities initially still boggles my mind.

Curation. 'Nuf said.

Interest-graph focus with an author/authority spin. Facebook owns social. Face it, that's going to be reality for a while. Attack Facebook's weaknesses rather than strengths: Real Names mandates, utter lack of privacy controls, morally bankrupt management, floods of irrelevant content. Unfortunately Google's been chasing those taillights instead of taking the winning move: flying the opposite direction at warp speed.

Attracting "thought leaders". Not sycophants (Mike Elgan and Jeff Jarvis come to mind). But people who are distinctly and genuinely original. Lower all barriers for them so long as the content is quality. And yes, I count Robert among the good (he's not Einstein, but his posts are interesting, informed, and he engages). This is the same recommendation / challenge I've presented Ello.[9]

Why listen to me?

I'm a space alien cat. I offer no claims to authority other than my own record, and the data and methods I present. Those last are independently verifiable, and I openly invite others to independently verify (again: kudos to Eric Enge who's actually done that).

If you don't care to believe me, fine. I'd suggest however that assume good faith[10]. My views and understanding are based on a mix of experience, on G+ and previously, which admittedly others can't verify, but also some substantial quantitative analysis, which is visible.

Draw your own conclusions.


1. Estimating G+ User Activity: 4-6 million active posters in January 2015 to date

This was the subject of a number of articles in January, 2015.

Eric Enge of Stone Temple Consulting did a far more rigorous follow-up:

2. "This One Trick Will Revolutionize Your Use of Social Media: Block fuckwits."

3. Robert Scoble, "Why Yo Daddy Won’t Use Google+: No Noise Control"

4. See: "Tracking the Conversation: "Mark Blyth""

"Tracking the Conversation: Postcapitalism"

The initial inspiration was a search for "Thomas Piketty" across several domains.

5. Why the first 36 and not the full 100? Because the queries are being run by an automated script (10,700 manual queries is a tad obsessive, even for me), and they're rate-throttled to keep from triggering blocks by Google. The process is now working through the 51st name. Still, the first 36 give a pretty good overall ndication.

While we're answering whys, why the FP Global 100 and not some other list with $YOUR_PREFERRED_INCLUSIONS? It's arbitrary, but as an extant list drawn from a fairly broad and representative interest group, it's a reasonable proxy for "where is conversation on substantive topics occurring?" Agree or disagree with any one member, if you're talking about them, you're not just talking about the weather (though you might be talking about the climate).

6. I didn't run the full FP100 query over the domain, but instead estimated counts using the top seven names, accounting for 66% of the hits for the FP36. Pope Francis was added to Pope Benedict given timeframe (Benedict's papacy ended prior to Ello's launch), which highly favours Ello (47% of total hits are Francis). Take with much salt. Ello's tiny, thatsthepoint.jpg.

7. For a discussion of strengths, see: "Ello As Blog / Publishing Platform"

Yes, changing the present site styling would be a major win. I've [written my own]( which I consider to be a massive improvement. I also have my own CSS for G+, whose stock UI is awful, and whose underlying CSS is an unmitigated clusterfuck. Google's use of Closure CSS minification is a category error.

A better Markdown parser (more complete, fewer bugs) is also high on my list.

8. Yeah, something else I've written about:

For what's wrong with G+ "Collections", admittedly a start though a bad one:

9. "This is a challenge to the Ello team (and Ello members)"

10. See Wikipedia:

Also: "Our social policies are not a suicide pact"

You're welcome to drop the assumption if you've found compelling reason to show it unwarranted. Though I've seen far too many attacks based on prior frames and lenses which my counterpary won't drop.
32 comments on original post
David Gerard's profile photoTony Sidaway's profile photoEdward Morbius's profile photoDmitry Lavrov's profile photo
+Tony Sidaway Reddit started as an enormous horde of fake accounts and fake votes(1), and now it's in all likelihood using fake accounts and fake votes to create an apparent support for idiotic ideology held by founders.

Add a comment...

David Gerard

Shared publicly  - 
Particularly after #oraclegate, I must note what a huge fan I am of Feminist Hacker Barbie.
God Emperor Lionel Lauer's profile photoDavid Gerard's profile photoChristopher Meid's profile photo
You mean B@rb!e? She rocks. 
Add a comment...

David Gerard

Shared publicly  - 
Peter da Silva's profile photoDavid Gerard's profile photoMirosław Baran (Jubal)'s profile photoGod Emperor Lionel Lauer's profile photo
Add a comment...

David Gerard

Shared publicly  - 
"What are you doing?" Steven asked Barbie. #oraclegate
Luke Seymour's profile photoMichael Olsen's profile photoDavid Ford (FirefighterBlu3)'s profile photoThomas Kehrenberg's profile photo
Add a comment...
Have him in circles
2,143 people
Daniele Zaini's profile photo
Imtiaz Ahmed's profile photo
George Fontes's profile photo
Ting Chen's profile photo
Jon Harald Søby's profile photo
Leela Mohan Peruboina's profile photo
Randal Bradakis's profile photo
Kristie Karaffa's profile photo
Arne Klempert's profile photo
So I just looked at download numbers, and it turns out AOO has knowingly distributed at least 8 million vulnerable copies of AOO since April 27. Rather than remove one file to fix it, they .. do nothing.
André Esteves's profile photoOtávio Cipriani's profile photoArild Ringøy's profile photoJose Pina Coelho's profile photo
CORRECTION: I originally wrote 50 million. That's since April 2014 - it's 8 million since April 2015. (And of course the 143 million before that.)
Add a comment...

David Gerard

Shared publicly  - 
How a Democrat invested in a Tea Party candidate, even lending a pollster, so she could beat him in the general election.
It was August 7, 2012, and I was standing in my hotel room in Kansas City about to shotgun a beer for the first time in my life. I had just made the biggest gamble of my political career—a $1.7 million gamble—and it had paid off. Running for reelection to the U.S. Senate as a Democrat from Missouri, I had successfully manipulated the Republican...
6 comments on original post
David “Q the Platypus” Formosa's profile photo
Add a comment...

David Gerard

Shared publicly  - 
tl;dr if you see anyone running AOO, GET THEM OFF IT AND ONTO LIBREOFFICE. AOO is a dead project, but the people squatting the name "OpenOffice" don't want to let go.

See also LWN discussion:
A couple of weeks I visited my mother back home in Norway. She had gotten a new laptop some time ago that my brother-in-law had set up for her. As usual when I come for a visit I was asked to look at some technical issues my mother was experiencing with her computer.
David Gerard's profile photoRob Shinn's profile photoMirosław Baran (Jubal)'s profile photo
I use LO right now, but nothing would preclude me from shifting to AOO if there were a compelling reason to do so. But there isn't.

Precisely. And that's a combined result of ASF/AOO obstinacy and more new developer and user oriented policy of TDF. I see the perceived licensing issue as a red herring.
Add a comment...

David Gerard

Shared publicly  - 
Postgres response to Oracle security statement: Please, security test our code!
God Emperor Lionel Lauer's profile photoDerek Pennycuff's profile photoGary Walker's profile photoSamuel Penning's profile photo
Sorry and sweet jab at Oracle. 
Add a comment...

David Gerard

Shared publicly  - 
I would be most interested to know what the numbers were for Buzz, before they unceremoniously shot it through the head.

G+ is fine, y'know. I'm still here. But goddamn, the fanboys' propensity for transparent bullshit about it ...
From a discussion with a reality-challenged, intellectually dishonest, Google+ fanboi

(Intended as a private archive post, though no harm in public sharing... See context note at bottom.)

OK, let's just address a few additional points.

On Experian/Hitwise

My "I'm accepting Experian/Hitwise's numbers on faith" is a simple statement that prima facie I'm accepting them. Not that my belief is unshakable, but that they seem reasonable, correspond with other data I've seen, including my own research into G+ activity numbers, public and private, and independent verification in whole or part of those numbers.


"Estimating G+ User Activity: 4-6 million active posters in January 2015 to date"

This looks at the most recent post date per profile for ~50k randomly[1] selected G+ profiles.

The results were largely independently substantiated by a subsequent, larger, and more systematic study by Eric Enge of Stone Temple Consulting.

"G+ Profile activity -- "active" public sharing accounts vs. "inactive" non-public-sharing accounts -- is there a silent but deadly active majority?"

Here rather than look at publicly visible posts, I looked at the "Followers" and "Views" counts on profiles themselves. While not directly related to posting activity, a view does require both followers and content. The results show far less exposure for profiles which don't show public visibility. 

Both approaches were replicated independently by Eric Enge of Stone Temple Consulting, providing a larger, more accurate, and detailed view of G+-specific activity. Enge's findings tend to support mine: G+ activity numbers are exceptionally modest given the 2 billion plus profiles "registered" with the service (2.2 billion at the time of my research, over 2.6 now).

"Hard Numbers for Public Posting Activity on Google Plus"

Note that both my and Enge's studies look only at G+, and don't compare activity to other sites. We've also been quite clear that the metrics we're using are idiosyncratic, but reflect accurately publicly visible activity data on G+.

More recently I've been looking at where discussions are happening, on specific topics, of particular interest to me. That is: I'm not representing that this is of interest to anyone else, but it does very much relate to what I am looking for. And, I might add, why I find G+ so disappointing. These do however provide for direct, cross-site comparison:


"Mark Blyth"

"Dissipative Systems"

General conclusion: if you're looking for discussion, on specific topics of interest (that is, not personal affairs), discussion sites, organised by interest rather than social fabric are much more useful.

Specifically: reddit, though a few other honorable mentions appear: Quora, Metafilter, SomethingAwful.

And blogs. If someone can figure out a blog comments system that works they'll be far more useful.

G+ itself fares poorly, though less badly relative to Facebook than total engagement numbers would indicate, in this regard. G+ Communities are absolute crap though. ***G+ discussions happen around individual posts, not Communities.***

Actually, for my purposes, G+ is better as public G+ discussions are actually accessible to non-users, while FB's frequently are not.

Your claim:

"It's an open secret (well, perhaps not for you) that statistics from Experian/Hitwise are not suitable to measure Social Traffic"

So, if this is an "open secret", ***Post your evidence that Experian/Hitwise are unreliable.***

I've searched: "experian hitwise (unreliable|biased|unsuitable)"

On DDG there are two results, neither of which address the claim.

Google turns up a few more results (9,390 presently), of which this Quora discussion tops the list:


Speaking specifically about UK Hitwise data (because I don't know enough about how they operate in other countries) - Hitwise data is based on ISP-level data from (they claim) 8 million Internet users.

This is probably the largest sample size of anyone providing these kinds of stats. As a result, Hitwise's strengths lie in number which make the most of this scale; for example; Analysis of how different search queries are driving traffic to a site, Analysis of days, or time of day, Share of market between different sites in the same category.  But, because they measure on a household level, they can't say much about actual users - age, gender etc. They can't see whether a household represents a single user, or a family of users.

So: yes, Hitwise doesn't give user-level granularity. But within its unit of analysis (household), data for multiple sites are reasonably comparable.

Or this overview of several web traffic tracking services:

Or Inside Story:

And if you're going to accuse Hitwise of a pro-Facebook bias, you'll have to explain this:
"A report by Hitwise shows that Facebook is losing market shares in the UK." 11 August 2011

Data published on Tuesday by Experian Hitwise prompted a flurry of headlines exclaiming “Facebook usage falls to three-year low” and “Facebook traffic is falling in the UK“. We know there are lies, damned lies and web statistics, but still – falling usage was not directly what the source data showed.

And, more to the point, as Alex has noted repeatedly, trending data should be accurate. The same long-term biases which affect G+ will likely reflect FB: Household composition, desktop vs. mobile, or other secular trends.

If you want to specifically address bias in Experian/Hitwise statistics, you must:

⚫ Indicate why there's significant variance among ISPs not captured in E/H's data and reports.
⚫ Indicate why there's significant variance among households not captured in E/H's data and reports.

Again: you've done none of this.

So apparently you've got nothing. I'll take your continued failure to address this as agreement.

As for my actions and reasons, I speak for them. Not you. Insisting that you do (likewise Daniel Well), is yet another display of intellectual dishonesty.

As I said, I've muted out of this conversation. No need to respondd, but I thought I'd expand on my brief comments.



1. As the piece notes, there's an assumption that the G+ sitemap files used are randomly filled with profiles. Given limited resources I tested this only cursorily, but the tests generally supported this: profiles within a given sitemap were created at widely differing times, had widely divergent names, and had location indicators all over the world in what appeared to be sensible patterns. Spot checks of a few of the 50,000 other sitemaps showed generally similar patterns. And the kicker: Stone Temple Consulting's much larger 500,000+ profile sample turned up sharply similar results. So the assumption appears to be well-founded.

From: / Sascha Pallenberg
Estimating G+ User Activity: 4-6 million active posters in January 2015 to date Trying to sort out what the actual active user count on G+ is is something of a cottage industry. I've suggested the label "Plussologists" (from Kremlinologists) for those so engaged. In his October, 2014 ReCode Interview, G+ bossman Dave Besbris rather pointedly said "I don't want to talk about numbers". Why not, Dave? Hiding something? This is an analysis w...
10 comments on original post
Add a comment...

David Gerard

Shared publicly  - 
Tony Sidaway's profile photo
Didn't work.
Add a comment...
Have him in circles
2,143 people
Daniele Zaini's profile photo
Imtiaz Ahmed's profile photo
George Fontes's profile photo
Ting Chen's profile photo
Jon Harald Søby's profile photo
Leela Mohan Peruboina's profile photo
Randal Bradakis's profile photo
Kristie Karaffa's profile photo
Arne Klempert's profile photo
  • Spartacus
Basic Information
Check the home page for quick bio.
Bragging rights
My modesty would rival Rupert Murdoch's.
  • Spartacus
Map of the places this user has livedMap of the places this user has livedMap of the places this user has lived