66 stories

Generate Hashtag-Based RSS Feeds for 7 Popular Mastodon Instances With Resstodon

1 Share

There’s a lot of hand wringing and analysis and punditry about what is happening and what might ultimately happen at Twitter. It’s interesting up to a point but my bigger concern is that Twitter as a content curation platform has turned into absolute garbage, at least for my topics of interest.

Many accounts I follow are gone or dormant. Many other accounts are locked down to avoid harassment or minimizing hashtags to avoid bot swarms. (I stopped using cryptocurrency-related keywords because I would get overwhelmed by reply bots and DMs. Eventually they started responding to non-hashtagged keywords and I locked my account entirely.)

Is this permanent? I don’t know and neither does anybody else. What I do know is this: Twitter is crumbling now and I need to figure out how I’m going to transfer my content curation work to Mastodon and whatever social networks arise. But I’m starting with Mastodon.

Mastodon is decentralized, which will make monitoring more difficult than a centralized resource like Twitter. On the other hand, Mastodon is more open in offering things like RSS feeds.

So RSS feeds are where I’m starting. And though it’s very early days yet, I’m doing a lot of experimenting and needed a way to create RSS feeds for Mastodon instances. So I took an hour and knocked up a very basic tool called Resstodon.

This is a screenshot of the landing screen for the ResearchBuzz search gizmo Resstodon.


Resstodon is a quick-and-simple way to generate RSS feeds for seven popular Mastodon instances:

  1. Enter a hashtag without the #.
  2. Click the button.

Resstodon will generate an OPML file with a hashtag-based RSS feed for each of the seven instances listed above and deposit it wherever your downloaded files go. OPML files are not executables, they’re specially-formatted text files. You can open it with a regular text editor if you want to examine it – it’ll look like this screenshot.

A screenshot of an OPML file as it appears in an average text editor. It's pretty much a jumble of unformatted text.


What OPML files are used for are importing multiple RSS feeds into a feed reader at one time. With Resstodon, you can generate an OPML file in a couple of seconds and import it into your RSS feed reader in a couple of clicks.

While I was testing this, I used Feedly to import the OPML file. It sometimes failed on some of the feed imports, but if I tried re-importing the feed it worked successfully (and avoided duplicating existing feeds.)

This isn’t even step one in my work with Mastodon and RSS feeds. It’s possibly step one-quarter. I’m going to generate some feeds and watch them a little, see how much duplication is taking place between feeds, then decide where to go from there. It might be that I want to focus on an easy external way to filter Mastodon RSS feeds, like IFTTT with custom JavaScript – that would be an easy thing to teach with provided templates. It might be that there’s enough overlap between feeds for common tags that I should instead concentrate on addressing a single RSS feed programmatically with intense filtering.

I have a huge backlogs of Gizmos I want to make, many involving Wikipedia, but I suspect I’m going to start putting a number of Mastodon and wider Fediverse projects there as well.

Read the whole story
449 days ago
Share this story

Paths to Putin

1 Share

The Trump-Russia investigation is not a “witch” hunt — it is a “which” hunt! [1]
  • Which organizations or people were key to connecting the Russians and the Americans during the 2016 US Presidential election subsequent transition?  
  • Which links/relationships may be the conduits of collaboration or collusion or conspiracy?  
  • Which network paths connected Trump to Putin? There was no direct connection between them prior to the 2017 Inauguration.
  • Which network paths were used to transfer information/money/data from one side to the other?
  • Which individuals and which paths were key to the whole operation?
We have gathered public data from investigative journalists and court documents filed in relation to the Trump-Russia 2016 election meddling inquiry. Our public data is a subset of that which government investigators (Special Counsel, Congress) have access to — yet, there is plenty of data for us to analyze.  We can visualize the data, and run various algorithms on the data, to see non-obvious patterns which may be hiding in the mass of connections.

Below is a map of 600+ nodes (people and organizations) that have been discovered to have business/political/personal relationships between Trump associates and Russia.  These nodes have over 1750 links/flows between them.  Nodes are colored by country of origin. A grey link connecting two nodes indicates one or more relationships between the two entities. Each link implies a two-way flow of communication.
Figure 1 – Trump and Putin Associates: People and Organizations
The two large nodes in the center of the map represent Trump (blue) and Putin (red).  The connections represent relationships/interactions in place before the 2017 Inauguration.  Trump and Putin had no verified meetings before the Inauguration. When two individuals are trying to keep their relationship covert, they will never establish a direct tie between themselves.  They will use trusted intermediaries to convey information/agreements or to pass money/resources between their two groups. This allows for “plausible deniability” (i.e., claims of “no collusion”) between the two parties.  Both Trump and Putin made sure the media knew that they had never met before 2017.

With all of the links in Figure 1 it is hard to understand the interaction patterns visually.  By running a simple network measure that looks at links within and between groups, we find that for both the Americans and Russians, they were linked mostly within their own group, but they had a significant number of ties to the other group!  This implies that the ties between the two groups were probably not accidental, nor random.  It was not a coincidence that these two groups were connected.  They had a purpose. 

We see similar patterns of interconnectivity in business organizations when two groups are working together on the same project.  The interaction patterns in Figure 1 are also similar to the communication flows between two business organizations that have recently gone through a merger. [2] Many interactions remain within the original organizations, but there is sufficient communication to the new partner organization for coordination and collaboration. If it looks like a duck, walks a duck and talks like a duck, it might just be a duck.  If it looks like a merger, talks like a merger, behaves like a merger, it might just be a merger!

How many Russians have interacted with Trump business and political associates before, during and immediately after the 2016 election?  Fifty (50) American individuals had links to sixty-five (65) Russian individuals creating many possible paths for information and resources to flow in the network.  These 115 individuals created 143 links/bridges across country borders (USA-RU). 

Figure 2 shows only those nodes(people) where a Russian interacted with an American.  The network in Figure 2 is a subset of the larger network in Figure 1.  There are only a few grey links in Figure 2 because we have removed the other nationalities and the connections within/inside each group that are seen in Figure 1. You don’t see the Russian-Russian, nor the American-American ties, just those connection between the two countries. 

Figure 2 – Which Trump associates had contact with which Russians?

The intersections (grey links) in Figure 2 are the bridges over which money / data / information / agreements/ instructions can flow — they are the conduits of cooperation / collusion / conspiracy.  Contrary to Rudy Giuliani’s claims, which blue node (American) in Figure 2 has the most connections with red nodes (Russians) ?

Network Paths
In human networks, it is not only the nodes(people) and links(relationships/flows) that matter, but the paths they form to move information and resources throughout the many locations in the network.  As we see in Figure 1, there is no direct link/path between Trump and Putin prior to the Inauguration, but there are hundreds of indirect paths from one leader to the other.  Indirect network paths have a length of two steps or more — there is at least one intermediary between the start and end of the path. 

A one step path is a direct connection between two friends, colleagues, or co-conspirators.  A – B  is a one step or a direct relationship.  Adding a link to A – B gives us A – B – C which is a two-step path between A and C.  We have the original one step path of A – B, and we have a second direct one-step path of B – C. Combined, these form a two step path, with B as the intermediary/connector.  Using this process we can build paths of any length.  A – B – C – D is a 3-step path involving four nodes. Count the links/dashes to get a quick idea of the length of a path.

In most of our networks, whether it is with colleagues or friends, we want to be as close to others as possible. We want many direct relationships, and if those are not possible, then we want to keep our network paths as short as possible.  Yet, in covert networks, the schemers do not want to have direct ties between the main parties.  They do not want to show an obvious and direct quid pro quo.  They want indirect paths so that they can have plausible deniability or intermediaries they can blame when a conspiracy is exposed.

Participants in covert networks look to build indirect quid pro quo [3] — paths of influence that are not obvious, and are harder to detect and prove.  In networks, distance (longer paths) can deceive, and that is an advantage to the builder of the covert network.  Unfortunately, distance also distorts and delays, so the covert schemer cannot build very long network paths to hide behind.  They must limit their key paths of trust to one or two intermediaries, or in rare cases three. In covert networks intermediaries are either trusted or threatened — problems of obedience and understanding amongst the intermediaries arise when the network paths get too long. In addition, a large number of people with some insight to the conspiracy is dangerous if the conspiracy is ever discovered.  The leaders of the conspiracy need to find a balance between long and short paths of communication to carry out their conspiracy.

The builders of the covert network will use more that one path to communicate, but they will probably limit the used paths to those that contain the most highly trusted intermediaries that have a history of prior communication.  Leaders need to trust their intermediaries, and the intermediaries need to trust each other and know how to best communicate. This usually requires some prior history between the two parties on each side of a link. The message flow will be faster, smoother and more accurate between people that have experience/context communicating with each other.  The Trump Tower meeting of June 9, 2016, is a good example of using trusted intermediaries via indirect paths to arrange/coordinate an important meeting.

Having the data of who is connected to whom, allows us to do all sorts of network analysis. [4]  We can find all of the indirect paths from Trump to Putin or vice versa.  Running our network algorithms we find over 500 indirect paths (with one to three intermediaries) between Trump and Putin.  Of course all of these paths are not used to communicate/conspire between the two sides, but it does give us an indicator of what is possible. With all of these possible paths of interaction between the two groups, communication is not a coincidence.  These 500+ indirect paths in Figure 1 contained many intermediaries — dozens of which showed up again and again.  Are the repeating intermediaries the most trusted links by both sides?  Possibly so. They were the connectors that were used repeatedly.  They were probably both trusted and well-located — the right person(s) in the right place at the right time.  Of the 500+ possible paths between Trump and Putin, less than 20% were actually utilized, or attempted — and of those, only 5% were relied upon. 

We looked at the collection of indirect paths and found 49 intermediaries that appeared most often on all network paths between Putin and Trump.  Those people show up in the map below.  Figure 3 shows just the network paths between Trump and Putin that contain these 49 key intermediaries. Figure 3 is a subset of the larger network in Figure 1 — showing the important indirect paths between Trump and Putin before the Inauguration. This map is colored the same as the first map — red nodes are Russians, blue are Americans, yellow nodes are Ukrainians and the green node is a European.

Figure 3 – Network Paths and Trusted Intermediaries Between Putin and Trump

Some may look at Figure 3 and ask: where are some of the people we have heard about in the news recently?  Where is Maria Butina?  Where is Roger Stone?  Where is Julian Assange?  They all showed up on various network paths, but they were not as well located, nor as often, as the 49 intermediaries that appear between Trump and Putin in Figure 3.  

  • Paul Manafort
  • Michael Flynn
  • Jared Kushner
  • Sergey Kislyak 
  • Michael D Cohen
  • Oleg Deripaska
  • D J Trump Jr
  • Dmytro Firtash
  • Viktor Yanukovych
  • Alexander Torshin
  • Rick Gates
  • Viktor Vekselberg
  • Roman Abramovich
  • Aras Agalarov
  • Ivanka Trump
  • Sergey Gorkov
  • Carter Page
  • Igor Krutoy

Outside of the two leaders, the best-connected bridges in Figure 3, sorted high to low, are as follows:

These 21 individuals are key in the network because they are on many paths of information flow — they form the best bridges in the network — they know what flows.  As is evident in Figure 3 and from the list above, Paul Manafort is on many key pathways between Trump and Putin.  He has sabotaged his own decision to cooperate with the federal investigators. What really happened with this critical node?

Roger Stone’s recent indictment [5] shows us his involvement in getting the hacked Democratic emails.  Stone’s main interaction was with Guccifer 2.0 (a cut-out for the Russian GRU) and direct and indirect contacts with WikiLeaks and Julian Assange, who were distributing the hacked emails shared by Unit 26165 and Unit 74455 of the Russian Intelligence Service: GRU.  Figure 4 shows who in addition to Roger Stone and Guccifer 2.0 possibly participated in this project. The highlighted lines in Figure 4 show the shortest possible path of communication.  Longer paths may also have been used to bridge both sides.
Figure 4 – Roger Stone as possible intermediary of hacked Democratic emails

Stone’s indictment was not for conspiracy. I expect (as do others) that Wikileaks and Julian Assange will be indicted next.  They will probably be indicted in a manner similar to the 12 GRU members.  If that happens, then maybe further/superseding indictments will be revealed for those who coordinated with Wikileaks to arrange the scheduling of further dumps of emails stolen by the GRU.


So, what have we learned from these maps of connections during the 2016 campaign?
  • Maps show the big picture of what is known about Trump-Russia interactions, before the 2017 Inauguration, based on data gathered by journalists and federal/state investigators. 
  • Denials by Trump and his associates that they had nothing to do with Russia before and during the 2016 election are visibly false.  Figure 2 makes that explicit.
  • Communication patterns between Trump & associates and Putin & associates match communication patterns/frequency found in many corporate mergers where two organizations are coming together and actively coordinating and working on common goals.  These two groups exhibit communication patterns of collaborating organizations.
  • If the interactions and communication between the Trump and Putin camps were just “normal business ties," why were they constantly denied, hidden, minimized, and lied about?  Patterns in the network reveal that this was not a normal business project.
  • Pattern of interaction shows typical pattern found in covert/corruption networks that is trying to hide its true intent. Leaders do not interact, while their underlings do. Leaders maintain unexpected distance and seek plausible deniability. Their aim is to execute an indirect quid pro quo. In a normal business project, leaders first interact to create agreement and set objectives, and then their staffs execute the project.  The leaders are never visibly excluded as participants in the project.  There is a direct tie between the leaders during the project.
  • In Figure 2 Trump has more connections with Russians than any one of his associates! 
  • Once he became the official Republican party candidate, and started receiving regular intelligence briefings, Trump was warned that the Russians may try to infiltrate his campaign. Trump did not report, nor seek assistance, with the many Russian direct and indirect interactions with his 2016 campaign. Trump kept praising Wikileaks though it had been already reported that the source of the Democratic emails was the Russian GRU.
As more data becomes available from court filings and trusted investigative reports, we will update our data, and our analysis, to track the changing dynamics of this covert network.  Stay tuned!
Read the whole story
1846 days ago
Share this story

Which Is Worse?

1 Share

In my nearly-17-year career with the federal government, I have now experienced shutdowns both as an “excepted” employee (required to work through it but not paid until it’s over) and a furloughed employee. A colleague asked me which one is harder, from my perspective.

They both suck. They suck differently, but they suck. Let’s break it down.

As an excepted employee, you must show up to work every day. Your back pay is guaranteed, but you won’t get paid for your work until the shutdown ends and payroll is processed. The civil servants who are excepted are often on front-line jobs: weather forecasters, NASA mission staff, TSA security workers, even the Coast Guard. Most of them serve public safety in some way or another. A shutdown asks them to work in a limited “business-as-usual” form, doing their primary functions but often barred from taking training, conducting research, working on special projects, or traveling.

Excepted employees are told that if they take leave, they will be placed in furlough status. Different agencies, managers, and shutdowns handle the status differently. Sometimes, the excepted employees are told they cannot return to work status once they take furlough; other times, they may flex between the two. Guidance in advance is never clear, and excepted employees are advised to cancel leave. Should they take their leave and move into furlough status, their predicament matches their furloughed colleagues.

Furloughed employees are not allowed to come to their buildings, check their emails, or conduct any work activities, even from home. They may not conduct work-related activities, including speaking at conferences or taking training, on their personal time if the work would have been a part of their “on the clock” duties. Their back pay is not guaranteed, though in every shutdown prior to the Great Shutdown of 2018-2019, their backpay has been approved. It does require an act passed by both the House and the Senate, then signed by the President. The behavior of both the recent Congress and the recent President has not always matched precedent, so until this bill is signed, furloughed employees live in a world of uncertainty about their pay.

Both excepted and furloughed employees carry added emotional and mental burdens. Excepted employees face working only some parts of their jobs, with inconsistent guidance about what they may or may not do while working. They feel unappreciated. They are frustrated because they are not eligible for unemployment due to their work status, nor do they typically have time to pursue other paying work because they are still reporting for their jobs.

Furloughed employees feel dismissed and marginalized, sometimes even by their own colleagues. They usually support mission-critical functions, though perhaps in a less immediate role than their excepted colleagues, but are signaled that their work is not important. They see their excepted colleagues receiving public accolades and food deliveries for their service, and they know they would be happy to be in their colleague’s shoes and helping and maybe eating a little more if they only were allowed. Many face steep work and learning curves when they return to duties that have piled up in their absence, causing stress both during the shutdown and afterwards that can linger for months or longer. Their uncertainty about pay status can be particularly troubling.

So, let’s be clear. Being excepted is not just “business as usual,” and being furloughed is not a “paid vacation.” They are burdensome and stressful, and both detract from the mission of the agencies and needs of the citizens that these civil servants serve. Nobody wins. Nobody is worse, and nobody is better. But, and this cannot be emphasized enough, we are all in it together.

Read the whole story
1861 days ago
Share this story

The Signal Network

1 Share

I spend a lot of time listening. A non-trivial act for me because my mind… it wanders. However, I’ve got a system. Feet flat on the ground, slightly clenched jaw, staring you straight in the eyes. I am full body listening. You have my complete attention. I am not missing a word.

We humans are experts at instinctively knowing where attention is focused. In a 1:1 situation, it’s clear from a sub-second glance at my watch to indicate to the other party that my focus is elsewhere. I am not listening. In that second, the quality of discourse plummets because the listening contract is broken.

My educated guess is that 50% of my job as a manager is information acquisition, assessment, and redistribution. It is my primary job and the efficiency with which I do this is a direct contribution to the velocity of the team.

Critical Freshness

In thinking about all the listening I’ve done and information I’ve acquired, I discovered I have a mental model for classifying information. It looks like this:

This grid has two axes. The vertical axis measures the criticality of a given piece of information. Critical information might look like:

  • Jake is about to quit.
  • The arrival rate of critical bugs is rapidly increasing.
  • A meeting just finished between Engineering and Sales where at each other’s throats. Nothing was resolved. Everyone left the meeting mad.

The horizontal axis is where this graph gets interesting. It measures freshness which is a synthetic measure of how long a given piece of information takes to get to the human who gets the most value from its arrival. Confused? Keep reading.

The interpretation of this graph is a very personal thing. You need to consider this graph as a thought experiment through a couple of different lenses. First, what is critical information to one human is irrelevant to another. Jake’s desire to quit is hugely important to his manager but less relevant to someone outside of the organization. Second, and worse, if the Jake information takes two weeks to get to Jake’s manager the information is not fresh and Jake’s manager has less time to take proper action.

Every single human in the organization has their version of this graph, and my thesis is the interpretation of this graph describes the health of your signal network.

Your Signal Network

Your signal network is the combination of all the information sources and all the information generated (or relayed) via those sources. The complete network is a combination of humans and robots, but for the sake of this article, let’s focus on human information sources. Back to the graph.

If you think about your average workday, you are continually discovering pieces of informational. Intentionally and accidentally. In meetings, in hallways, and in the cafe. Your working life is chock full of rapidly arriving information, and your brain must quickly digest, parse, pattern match, and make a judgment regarding each piece of information. What is it? How critical is it? What am I going to do with it? Should I pass it along? And to whom?

Each quadrant of this graph describes a different assessment of a piece of information. Let’s walk through each:

  • Stale & Slow The lower left quadrant is the most boring quadrant. The information here is not relevant and isn’t fresh, but who cares? It’s low signal information, and it’s stale, so there is no need to act.
  • Voluminous Spam The lower right quadrant is less annoying. You’re still dealing with less critical information, but the more you move to the right, the fresher the information. I’m sure learning lots of useless things quickly. At an extreme, it’s spam. An organization spends energy moving information hither and fro. If you’re seeing a lot of information falling into this quadrant I am concerned about the overall efficiency of your team. If you’re seeing a lot of useless information on a day to day basis, what about the rest of your team? How much time is the team wading through the noise to find signal? How much time are they wasting looking for nuggets of relevancy?1
  • Critically Fresh The upper right quadrant is your informational sweet spot. Critical information is getting to you in a timely fashion. Yes, it’d be super if all the information was further up and to the right, but the fact the information is in this quadrant is a win. The vibe here is a distinct lack of surprises. When a piece of information lands on your plate, it’s fresh. It’s clear someone just made this horrible decision, and you have ample time to coach them in the correct direction.
  • Important, But Slow The final quadrant, the upper left, is the danger zone. Critical information arriving slowly to the humans who need it the most is the source of much of your organizational consternation, and I need a whole section to explain why.

No Surprises

Members Only was the code name of one of my managers from The Start-up You’ve Never Head Of, and he exhibited many classic management tropes. We didn’t know what he did all day, he never scheduled 1:1s and when they were scheduled they were often rescheduled into non-existence without notice, and when you finally got him pinned down in a meeting room, when you cornered him with the vital issue, he leaped to the first conclusion that crossed his mind and stated it as a fact.2

Members Only was fond of simple, pithy management proclamations which made a lasting impression on me. He told me one the first week he joined, “No surprises.”

My interpretation of these two words from Members Only was not generous. What I heard was, “Make sure I know what’s up, so I don’t look bad.” What he might have meant (but I’ll never know) was, “Make sure our team of humans has the best information as quickly as possible so they can make the best decisions as quickly as possible.”

Information consistently falling into Important, but Slow means there are surprises. You’re discovering unexpected developments occurring on the team long after they happened. You’re unable to react because the time to act has passed. The conclusion is already history.

You can spend a lot of time and money investing in processes, tools, and architects that you believe are necessary to critical and timely information flowing, but where I consistently invest is in the team. I demonstrate to the humans the value of effectively detecting, assessing, routing, and retransmitting information across the organization.

High Signal Humans

I have an internal measure which grades the following on any given day: how much critical information did I discover? And how fresh is it? In a rapidly growing organization of humans, the volume of new information created daily increases daily.

If you buy that the healthy flow of information is an essential practice as a leader, then you understand why I religiously hold 1:1s. It’s a regular meeting where I make it clear what critical information I care about and where I consistently share the critical information my team needs. It’s never a perfect transaction. I often incorrectly flag essential information that is spam, you will, too. Over time, we will calibrate. In time, we won’t wait until the 1:1 to relay information because we’ll intuitively understand that for this given piece of information, the faster it lands in the right hands, the higher the value.

Your ability to effectively lead is a function of the collective quality of the decisions you make on a daily basis. You can take your time on many decisions. You can wait days or weeks until you’ve gathered all the relevant signal necessary. Other decisions must be made right now. At that moment, the health of your signal network, the amount of critical information that has arrived in a timely fashion makes the difference between an informed decision and the flipping of a coin.

The health of your signal network is one lens into the health of your team. Critical information freely moving around the organization decreases surprises, improves the quality of decisions, and builds trust. They are your signal network. And you theirs.

    1. Gossip. Briefly. Some information that shows up is gossip. Its half-informed opinion relayed as fact. Gossip is a trigger for me because it’s often a precursor to the worst kind of politics, but gossip is signal. Rather than becoming angry, rather than wasting time on figuring out “Who would say that?”, I choose the dissect the gossip: What perfectly reasonable question is being asked with the inflammatory chunk of gossip? What observation is being made? This approach does not always work. ↩
  1. Much of the initial material for Managing Humans was created during this period. ↩
Read the whole story
1894 days ago
Share this story

Using a Google Sheet to Build Search Queries for Unknown Topics

1 Share

I spent a lot of time building tools at work. There were things we needed to automate, but I didn’t have a server or any kind of programming platform to put things on. Because of that I discovered you could do a lot just using Google Sheets.

I built a tool that let you scan in ISBNs and get the title and author of a book. I built another that let you put in the title of a movie and get a plot summary and list of actors. I designed another one that let you scan in a CD and get recommended genres and sub-genres. (That one I had to hire someone to program as the implementation was a bit beyond me.) There’s no telling how much time those tools saved me at work, and they left me with a slight mania for making useful spreadsheets, though it’s not really one of my job functions anymore.

Last week I got an e-mail from a reader named Joe asking for help researching a topic he didn’t know a lot about. Then at about the same time, I read about an add-on that would let you import Wikipedia data into Google Sheets. And I thought, “Hey, this might be a nifty way to quickly build Google search queries when you don’t have a lot of information.”

And so the Web Search Query Builder 5 Million Google Sheet was born. (The “5 Million” part is in honor of the Hydraulic Press Channel, which seems to name all its original tools the dup-de-dup 5 Million. Visit YouTube to see the Rock Maker 6 Million (it’s an upgrade) turn hair into powder.)

You can try the sheet at https://docs.google.com/spreadsheets/d/1hl0Ku9eqLIcr3piiz5DI_bbeZnQhZP-4eMLRBFLWJ44/edit?usp=sharing . I encourage you to make a copy of it for yourself and play with it. I’ve found it’s useful for getting a quick overview when I run into an unknown concept or name while doing ResearchBuzz; it’s much faster than searching for it “cold,” with no contextual hints. (And since I run into topics and names I don’t know constantly, I’m hoping this will save me some considerable time.)

PLEASE NOTE VERY IMPORTANT: If you do make a copy of it and (and as I said before, please feel free) you will need the Wikipedia and Wikidata Tools add-on for Google Sheets. It’s free and you can find it here: https://chrome.google.com/webstore/detail/wikipedia-and-wikidata-to/aiilcelhmpllcgkhhpifagfehbddkdfp?hl=en 

In this article I’ll give you an overview of the sheet and explain how the tabs work.

What the Heck Is This

What this the heck is, is a sheet that uses the Wikipedia and Wikidata Tools add on to draw data from Wikipedia in response to a query and then build from that data a set of Google search queries which are tweaked slightly by domain limitations and thematic keywords. In the case of this particular sheet, I’m including limitations to the .edu and .gov domains, and adding additional keywords with the hope of finding more teacher-friendly resources. The tool is set up in a series of five tabs.

Tab 1 – Queries

screenshot from 2018 05 08 17 48 41


The first tab is for entering your search term. General searches like “cow” will get you so many results as to be useless. I find that names and events and somewhat obscure concepts are the most fruitful; these searches for example brought good results:

Great Patriotic War
Bangladesh Liberation War
Mzwakhe Mbuli

Underneath the query box you’ll see count results for your query; enter your search term, and these boxes will briefly reset to 1 and then update with the number of queries returned, thesaurus lookup results, outbound link topic results, and category results. There’s a maximum of 500 items per line. If you find that you’re getting 500s or 100+ for everything, your search may be too broad. I do find, though, that popular culture figures tend to get very high numbers no matter what because they’re linked to so many things.

Once you enter a search term, all the other tabs will update.

Tab 2 – Wikipedia Data

This tab displays all the results that are pulled from Wikipedia and put in this sheet. The data here will be used in building search query URLs in the following sheets. I find that glancing at just this sheet can give me a general overview (very general) of a name or concept.

screenshot from 2018 05 08 19 47 26

Tab 3 – Search Engine queries

This tab presents you with a number of Google search URLs. The first set uses the first five lookup keywords to create general Google searches. The second set uses three keywords at a time to create a more specific query. (It works like this: the first URL incorporates the 1st, 2nd, and 3rd row of Wikipedia query results, the second incorporates the 2nd, 3rd, and 4th row, etc.)

The third set uses five keywords at a time. That one can be so specific as to give you no results!

Hold your mouse pointer over a URL and you’ll get a little pop-up box with the option to open the URL (you’ll see what that looks like in the screen shot.) Unfortunately the URLs aren’t really directly clickable; that’s a Google interface decision and not something I can change.

(Also note that if you run a search that has a limited number of results from Wikipedia, all of the search URLs won’t populate with keywords. They’ll still be on the page, they’ll just be empty.)

screenshot from 2018 05 09 05 09 16

Tab 4 – Search Engine TLD-Domain Restricted Links

Tab 4 takes the same kind of searches done in tab 3, but adds domain limiters so the first set queries is restricted to .edu domains only and the second set is restricted to .gov. There are simple and medium queries here, but the “deep dive” queries of five keywords at a time are removed. I found doing searches that large AND restricting by top-level domain tended to produce few to no results.

screenshot from 2018 05 09 05 09 25

Tab 5 – Google Queries for Finding Learning/Reference

This set of Google queries removes the domain restriction but adds a couple of keywords designed to get the results to go in a more scholarly/learning direction. This isn’t perfect and the kind of result you’ll get will depend a lot on your initial search.

screenshot from 2018 05 09 05 09 35

This Could Go a Lot Further

When I initially developed this, I didn’t know how useful I would find it. But I showed it to my friend Kathy Jacobs, who looked at it and very helpfully broke it in a couple of places (when you make things like this, it’s invaluable to have someone technical enough to poke at it, break it, and then explain clearly to you how they broke it) and she liked it a lot. So maybe you will too.

Obviously this could go a lot further. I could add an option for specifying domains (top level or otherwise) that you want to restrict your generated searches to. You could specify your own keywords. Maybe customizable country restrictions for your searches.

Let me know if you find this useful and would like to see some upgrades. I’ve only done a little spreadsheet work since my job changed, and I realize I miss it!

Thanks to my great patrons at Patreon who are supporting ResearchBuzz and affording me time to write these articles.

Read the whole story
2113 days ago
Share this story

How to Rands


Hi, welcome to the team. I’m so glad you are here at $COMPANY.

It’s going to take a solid quarter to figure this place out. I understand the importance of first impressions, and I know you want to get a check in the win column, but this is a complex place full of equally complex humans. Take your time, meet everyone, go to every meeting, write things down, and ask all the questions – especially about all those baffling acronyms and emoji.

One of the working relationships we need to define is ours. The following is a user guide for me and how I work. It captures what you can expect out of the average week, how I like to work, my north star principles, and some of my, uh, nuance. My intent is to accelerate our working relationship with this document.1

Our Average Week

We’ll have a 1:1 every week for at least 30 minutes no matter what. This meeting discusses topics of substance, not updates. I’ve created a private Slack channel for the two us of to capture future topics for our 1:1s. When you or I think of a topic, we dump it in that channel.

We’ll have a staff meeting with your peers every week for 60 minutes no matter what. Unlike 1:1s, we have a shared document which captures agenda topics for the entire team. Similar to 1:1s, we aren’t discussing status at this meeting, but issues of substance that affect the whole team.

You can Slack me 24 hours a day. I like responding quickly.

If I am traveling, I will give you notice of said travel in advance. All our meetings still occur albeit with time zone considerations.

I work a bit on the weekends. This is my choice. I do not expect that you are going to work on the weekend. I might Slack you things, but unless the thing says URGENT, it can always wait until work begins for you on Monday.

North Star Principles

Humans first. I believe that happy, informed, and productive humans build fantastic product. I optimize for the humans. Other leaders will maximize the business, the technology, or any other number of important facets. Ideological diversity is key to an effective team. All perspectives are relevant, and we need all these leaders, but my bias is towards building productive humans.

Leadership comes from everywhere. My wife likes to remind me that I hated meetings for the first ten years of my professional career. She’s right. I’ve wasted a lot of time in poorly run meetings by bad managers. As an engineer, I remain skeptical of managers even as a manager. While I believe managers are an essential part of a scaling organization, I don’t believe they have a monopoly on leadership, and I work hard to build other constructs and opportunities in our teams for non-managers to lead.

I see things as systems. I reduce all complex things (including humans) into systems. I think in flowcharts. I take great joy in attempting to understand how these systems and flowcharts all fit together. When I see large or small inefficiencies in systems, I’d like to fix them with your help.

It is important to me that humans are treated fairly. I believe that most humans are trying to to do the right thing, but unconscious bias leads them astray. I work hard to understand and address my biases because I understand their ability to create inequity.

I heavily bias towards action. Long meetings where we are endlessly debating potential directions are often valuable, but I believe starting is the best way to begin learning and make progress. This is not always the correct strategy. This strategy annoys those who like to debate.

I believe in the compounding awesomeness of continually fixing small things. I believe quality assurance is everyone’s responsibility and there are bugs to be fixed everywhere… all the time.

I start with an assumption of positive intent for all involved. This has worked out well for me over my career.

Feedback Protocol

I firmly believe that feedback is at the core of building trust and respect in a team.

At $COMPANY, there is a formal feedback cycle which occurs twice a year. The first time we go through this cycle, I’ll draft a proposed set of goals for you for the next review period. These are not product or technology goals; these are professional growth goals for you. I’ll send you these draft goals as well as upward feedback from your team before we meet so you can review beforehand.

In our face-to-face meeting, we’ll discuss and agree on your goals for the next period, and I’ll ask for feedback on my performance. At our following review, the process differs thusly: I’ll review you against our prior goals, and I’ll introduce new goals (if necessary). Rinse and repeat.

Review periods are not the only time we’ll exchange feedback. This will be a recurring topic in our 1:1s. I am going to ask you for feedback in 1:1s regularly. I am never going to stop doing this no matter how many times you say you have no feedback for me.

Disagreement is feedback and the sooner we learn how to efficiently disagree with each other, the sooner we’ll trust and respect each other more. Ideas don’t get better with agreement.

Meeting Protocol

I go to a lot of meetings. I deliberately run with my calendar publicly visible. If you have a question about a meeting on my calendar, ask me. If a meeting is private or confidential, it’s title and attendees will be hidden from your view. The vast majority of my meetings are neither private nor confidential.

My definition of a meeting includes an agenda and/or intended purpose, the appropriate amount of productive attendees, and a responsible party running the meeting to a schedule. If I am attending a meeting, I’d prefer starting on time. If I am running a meeting, I will start that meeting on time.

If you send me a presentation deck a reasonable amount of time before a meeting, I will read it before the meeting and will have my questions at the ready. If I haven’t read the deck, I will tell you.

If a meeting completes its intended purpose before it’s scheduled to end, let’s give the time back to everyone. If it’s clear the intended goal won’t be achieved in the allotted time, let’s stop the meeting before time is up and determine how to finish the meeting later.

Nuance and Errata

I am an introvert and that means that prolonged exposure to humans is exhausting for me. Weird, huh? Meetings with three of us are perfect, three to eight are ok, and more than eight you will find that I am strangely quiet. Do not confuse my quiet with lack of engagement.

When the 1:1 feels over, and there is remaining time I always have a couple of meaty topics to discuss. This is brainstorming, and the issues are usually front-of-mind hard topics that I am processing. It might feel like we’re shooting the shit, but we’re doing real work.

When I ask you to do something that feels poorly defined you should ask me for both clarification and a call on importance. I might still be brainstorming. These questions can save everyone a lot of time.

Ask assertive versus tell assertive. When you need to ask me to do something, ask me. I respond incredibly well to ask assertiveness (“Rands, can you help with X?”). I respond poorly to being told what to do (“Rands, do X.”) I have been this way since I was a kid and I probably need therapy.

I can be hyperbolic but it’s almost always because I am excited about the topic. I also swear sometimes. Sorry.

If I am on my phone during a meeting for more than 30 seconds, say something. My attention wanders.

Humans stating opinions as facts are a trigger for me.

Humans who gossip are a trigger for me.

I am not writing about you. I’ve been writing a blog for a long time and continue to write. While the topics might spring from recent events, the humans involved in the writing are always made up. I am not writing about you. I write all the time.

This document is a living breathing thing and likely incomplete. I will update it frequently and would appreciate your feedback.

  1. Speculation: there is an idea in this document that you’d like your manager to do. Thesis: Just because I have a practice or a belief doesn’t mean it’s the right practice or belief for your manager. Suggestion: Ask your manager if they think my practice or belief is a good idea and see what happens. 
Read the whole story
2176 days ago
Share this story
Next Page of Stories