A citation to a live website can look complete until the page disappears. The World Factbook shutdown removed the original browsing path, redirected familiar addresses, and left researchers dependent on preserved copies. We needed a method that could survive another archive moving, changing its interface, or losing an individual capture.

The first problem after The World Factbook disappeared was deciding what an archived entry could still prove. A preserved page doesn't become unreliable merely because the live site closed. It does become fixed in time, and that limit matters differently for every field.

The CIA announced yesterday that The World Factbook has been discontinued. The announcement described the publication's history, from a classified reference launched in 1962 to an unclassified edition, a public website, and a resource used by researchers, teachers, journalists, travelers, and government readers. It didn't give a reason for ending the publication.

Foot pursuits and vehicle pursuits need the same dramatic decisions, but they don't operate at the same speed or over the same distances. Counting squares works for a short tactical movement. It becomes awkward across stairwells, traffic, alleys, highways, crowds, and changing routes.

The SRD's Influence action wasn't deep enough for a game centered on access, disclosure, authority, leverage, and human judgment. A single attitude and check can resolve a simple request. Espionage play regularly asks harder questions: Does this person know the answer? Can this person grant access? What would compliance risk? Which obligations or limits control the response?

• We spent a lot of time discussing attribution. The members of our group voted to not be credited. They’re mostly not gamers, they’re our friends who play D&D with us and agreed to help playtest this. A few have negative views of the online roleplaying community. They also don’t feel that they’ve contributed anything, although they very much have.

Research, surveillance, technical work, and other sustained tasks create a recurring rules problem. One roll can make hours of work feel weightless. A string of identical rolls creates the opposite problem, since the group keeps rolling until enough successes appear or patience runs out.

The standard ability-and-skill format is compact, but fixed pairings can give the skill list more authority than the fictional method. Modern operations expose that weakness quickly. Intimidating someone, treating an injury, entering a building, or examining a security system can depend on very different aptitudes depending on what the agent actually does.

• The equipment section is mostly done at this point. the decision to not go for specific weapons (Beretta 92, Glock 19, etc.) but stick to categories (pocket pistol, service pistol, heavy pistol, etc.) helped. In game terms is make very little difference, although there will certainly be people who not only dislike out chosen categories but the statistics assigned. Which is fine. The game will be open source, so they can write their own alternative statistics.

The familiar d20 mechanic gave us a useful starting point for OSINT-6 CORE RULEBOOK. Familiarity reduces the amount of basic procedure a group has to learn. It also creates a problem: a familiar roll can become automatic. Someone names a skill, rolls, and waits for the Game Master to decide what the result means.

E6 gave us three durable ideas: six levels can sustain a full campaign, capable characters can remain mortal, and planning can matter more than continued numerical escalation. It didn't provide a ready-made advancement chapter for OSINT-6.

The six-level limit in the OSINT-6 Core Rulebook draws on E6, a form of d20 fantasy that uses levels 1 through 6 for the full campaign. The useful idea wasn't that every E6 rule belonged in a modern game. It was that a short level range could support capable characters without turning continued advancement into a change of scale.

Mophidius has launched a Kickstarter for The Spy Game, a 5e-based espionage roleplaying game. Initially our hearts sank, but to be honest, it’s surprising it hasn’t already been done. And the existence of one d20-based espionage game doesn’t mean we can’t continue developing ours. There are, after all, how many 5e fantasy games when D&D has and will always have the majority market share?

SRD 5.2.1 is a fantasy roleplaying document. Its rules assume heroic classes, species, spells, magic items, monsters, treasure, and advancement through 20 levels. Those assumptions aren't confined to isolated chapters. They shape healing, travel, equipment, combat, social interaction, character growth, and the kinds of problems the rules expect players to solve.

The strongest argument for using SRD 5.2.1 wasn't popularity by itself. A familiar chassis is useful when its parts continue to solve the new game's problems. Several parts did.

The first major rules decision for the OSINT-6 Core Rulebook was whether to build a new engine or adapt an established one. We chose the System Reference Document 5.2.1 as the foundation because it offered a familiar resolution language, a complete body of usable procedures, and clear publication under the Creative Commons Attribution 4.0 International License.

• We text fixed hit point in the last session. Instead of rolling, each character gets (3 + CON bonus)*level. For example, a 2nd level character with a +1 con bonus has 8 hit points. The players all replace the current rolled hit points with the fixed amount. There was very little different. In a grounded, E6-inspirated game, if you get shot you’re messed up and might die, and if you get shot a lot you will almost definitely die. It suits the genre and removes a step from both character creation and levelling up that doesn’t really add anything. We also text fixed amounts for recovery and healing, but we’ll need to play with that so injured characters don’t either walk off a gunshot wound after a long rest or linger with an injury for an amount of time that effectively takes the character out of play.

The first thing the playtest group agreed on was no James Bond. As we flesh out the rules, particularly the equipment list, we won’t be including any imaginary gadgets, card enhancements, or fantastic versions of real-world devices. Keep it grounded. If we do release this as fully open source and someone wants that, they can create it themselves. Maybe someday we’ll do a supplement with over-the-top rules, but I think that’s as likely as releasing a sourcebook for levels 7-20.

We’re dropped the idea of using hero points. I think we were thinking of the James Bond RPG, which was arguably the first RPG to use them, and unconsciously trying to emulate it. They’re an optional rule in the 5e DMG but I don’t think they’re in the SRD, so we’d have to create our own mechanics for it.

But it doesn’t fit with what we’re doing. It would erase some of the point of capping advancement at 6th level to prevent power creep. And Alec Leamas couldn’t spend a point to keep Liz (or Nan) from getting shot while going over the wall.

Playtesting with the rough draft will begin when everyone is available again in the Autumn. Until then we’re writing and researching. The general consensus is that if we’re capping advancement at level 6, we should go with milestone-based advancement instead of awarding experience points. It makes completing missions successfully feel like the point, instead of doing individual things along the way like killing every minor bad guy or stealing everything that’s not nailed down in order to gain XP. This is definitely the right focus. We’ll figure out something like how many missions it takes to advance for each level, and how many missions you need to complete after level 6 to gain a new feat.