Grams’ Admin

Figure 71: FreeFor forum that uses IRC

Due in part to Coleman, the Anonymous whisperer, along with subsequent interventions, the reputation of Anonymous has shifted from criminal to almost heroic in regard to some of their actions involving ISIS. ‘Anonymous might be no friend to the US government: but it recently

“declared war” on ISIS. In this space, governments would be wise to work with anyone willing to help’ (Bartlett 2014b:12). Furthermore Bartlett’s ethnographic research of the Dark Net shows that ‘it was Anonymous hacktivists (and then more recently two vigilante hackers)––who helped to identify and remove Tor Hidden Service sites which were hosting child pornography’

(ibid). Since 2014 there have been a range of ‘operations’ identified by hashtags including

#Cyberprivacy, as well as the targeting of terrorists and corrupt and racist organisations, such as the 2020 hack and the edited release of over one million police files by DDoSecrets,

‘BlueLeaks’, in the wake of the murder of George Floyd in 2020. However, the shifting identities of Anonymous members, as well as their activities on the Dark Net, remain hidden.

as useful and easy to use as the clearnet’ (Anonymous 2014). Already in 2014, the admin shares how specific keywords relating to ‘trending’ topics were added to Grams, mimicking Google’s business model as explained in Chapter 5:

I’m conflicted by the need to monetize the site, while also creating features that will better the user experience. The next major feature, I think, will do both. I’m calling it

‘Gramwords’, and basing it the Google Adwords system. Now that the weighted keyword search algorithm is in place, I can really sink my teeth into this project.

Vendors and Tor site owners alike will be able to purchase sponsored keyword listings for their sites and/or products (Anonymous 2014).

Figure 72: A ‘domain seized’ image of deepdotweb. (August 2020)

At that time, Grams was designed to have the look and feel of Google, with a Google inspired colour palette and other Google-like features including the infamous ‘I Feel Lucky’, which acts as a type of filter to let the user sort the results in different ways: relevance, price and date.

Connecting back to Chapter 5, Grams anonymous admin offered some insight into the intentions and the developments of the search engine:

The big feature I’ve been working on for a while and finally finished this week was to overhaul the search algorithm. Once I started working on an ad-words feature, I realized that the search algorithm needed to be more keyword based since the previous algorithm only searched the database using full-text method. At the same time, I realized that a lot of searches were for keywords not necessarily related to the marketplaces, which meant I needed a full Tor search engine. I first created a full Tor search engine using weighted

keywords, which analyses webpages based on the number and placement of keywords on a page. After that was done, I created a similar system to find weighted keywords in the Dark Net market listings. Once both systems were completed I merged the database systems so that users will now see Tor sites and Dark Net market listings side by side in the search results. Search results will now be more relevant, accurate and diverse. The system is not perfect yet- I will be tweaking the different weights given to keywords over the next couple weeks, to give users the best results possible (ibid).

Figure 73: Interview Grams admin on the now defunct ‘deepdotweb.com’. Warning: Your Tor Usage is Being Watched

When I first found the site I received a warning and the banner ad to the right, as I was not using TAILS but only Tor at that moment. (Figure 73) I then used TAILS and discovered a special feature of Grams, ‘Flow’, which gives an overview of specific keywords in alphabetical order that when clicked upon, access Hidden Service sites as an alternative to remembering onion addresses. (Figure 74)

When I clicked on the first keyword of Gram’s Flow, ‘alpha’, this led me to the Alphabay site, where BitTorrent sites such as the Pirate Bay and information repositories on different ways to perpetrate credit card fraud are also prevalent, besides the drugs, guns and ‘rent-a-hacker’

services.¹³¹ (Figure 75) With the keyword ‘Rent-a-Hacker’ I received this login. However, I was redirected to the AlphaBay validation again before I could proceed and I decided not to

continue. (Figure 76)

131 As much as Bittorrents were predominant in the earlier days of Tor, there has been much backlash inside the community as the traffic slows down Tor too much and inhibits the TorFlow.

Figure 74: Flow offers keywords for the user to click on

Figure 75: entrance to the AlphaBay marketplace

Figure 76: Rent a Hacker redirect

Although ‘Grams has recently introduced trending searches and vendors can even buy sponsored keyword listings space for their sites and products’ (Bartlett 2014:162), unlike Google’s Adwords system, the goal is not to collect all of the users’ data and direct targeted advertising. Here the user is not personalised, with little information collected about the users who are anonymous and use pseudonyms and encryption keys, as discussed in Chapter 6. The person wanting to advertise makes an ad and determines which keywords are relevant and how much they are willing to pay a day, much like ‘affiliate advertising’. Users are assigned a Bitcoin address, where they need to add Bitcoins on regular basis, so the ‘Gramword’ system can function. Child pornography is not allowed on any banner ads, in line with Grams’ terms of usage.

Also, in contrast to Google, Grams allows any Tor site to be indexed and the user has the ability to ‘Add a Site’ so that they can add their own site to the index. ‘Tor site owners will get the best results on Grams by making sure their metatags such as title and description are accurate, just as they would SEO for a clearnet site’ (Anonymous 2014). Ultimately the goal of the Grams developer is to create an up-to-date search engine that offers relevant results for the seeker, only each user does not have a Grams profile. Intentions are clear––the admin hopes to make Grams the most trusted, reliable and popular search engine on the Dark Net. Many sites on Tor’s Hidden Service adopted the search engine Grams’ specially developed API and there is a subreddit where community members publish feedback that keeps up to date reports on which sites are available, along with providing vendor feedback.¹³² Although not having purchased anything on the Dark Net, I explored giving feedback, by reading comments on the Grams subreddit, such as this one:

132 https://www.reddit.com/r/onions/comments/22z3qe/grams_beta_version/

Hello, Grams scrape feedbacks from the markets sir, you must sign up to your market, buy some stuff and then place a feedback before you will see it on Grams. If you want to add a Grams review, you must search for your vendor on the InfoDesk, then click on the

‘review’ button and place it. That’s how it works :).(Figure 77)

The numbers in parentheses shown in the image are reviews, not sales and Grams plans to have a vendor directory on the site, where reviews will be collated from different markets.¹³³

Updating the site every three days, they added new features to the search algorithm to give the user the best results possible, such as a ‘scoring system’ based on ‘keywords, number of transactions, good reviews, and the number of clicks from our site’ (Anonymous 2014). The Grams anonymous admin also planned to add other features for discrimination such as price, country and market. When this interview was conducted, sites like Agora, Pandora and Silk Road 2 were still up and running and they were looking to integrate more sites into the search engine.

Figure 77: Grams Info Desk