Knowledge kills. Knowledge evil.
I think granite is your best bet (or something similar; geochemists can suggest the ideal mineral). Physically and chemically pretty tough and inert – a bit more brittle than metal, but won’t oxidize or otherwise react with much. Can be carved reasonably small.
Not valuable for anything other than bulk building material (make the monoliths triangular and people will be less likely to try and make walls out of them; that, or make them too big for six men to lift, so if they are ever moved then it will be a big enough deal that everyone will pay attention to the thing).
Of course, a slab of granite exposed to sandstorms will get ground down fairly quickly to the point where inscriptions will be hard to read. But I don’t know anything that wouldn’t. I think the only way to get your message to future sand-dwellers is to put it on the bottom of a tall obelisk and then bury it halfway. Eventually, someone will wonder what this thing sticking out of the sand is, and start digging it out.
If intelligible words are important, then best bet is to inscribe the same message in three different languages on each monolith. Worked for hieroglyphics. Make a list of say 12 widespread languagesand pick three for any given monolith.
But repurposing it doesn’t erase it, and might even help preserve it. Many tablets, stelas, etc… have been found and reconstituted by archaeologists because their parts had been reused in wall fillings, foundations, etc… while they might otherwise have eroded or been completely dispersed. So, it’s not necessarily a bad thing.
When I think of it, two carved stones (one depicting a face, another what seemed to be a wolf), I would guess dating back from the middle age, were apparent in a relative’s old house’s wall, I guess they originally belonged to some religious building.
As long as the scavengers decide to use the blocks whole, and not break them up to fill in the spaces they need in their wall.
If I remember right, the most easily eroded rock is halite (salt) while the hardest is laterite (Fe3O4.) Ankor Wat is made of laterite bricks. The pyramids Gizah won’t last 4,000 years (qualified) in a tropical setting.
Well, precisely I was thinking of examples of stelas that had been broken up and used as filling. Archaeologists end up with a puzzle, but a well preserved puzzle.
Though when I think of it, if your civilization is advanced enough to have archaeologists fumbling around broken tablets, maybe their content isn’t as important anymore.
On the other hand, if your civilization isn’t advanced enough, you probably won’t be able to decipher the blocks, regardless how well preserved and easy to find they are.
I guess the “blocks” should be made numerous enough and attractive enough to be well known and intriguing, so that people would think hard about them early on. This would require to make them even more resilient since they would then probably be handled, moved, etc.. a lot.
Alternatively, the goal might be at the contrary to make them inconspicuous enough to be noticed only when a civilization is advanced enough to be able to access to them, study them and decipher them (“how comes we always get those small bits of corundum when we drill for oil?” “Why is there a black monolith on the moon?”)
Sorry for this collection of random thoughts.
Artificial sapphire is available in fairly large sizes - not quite the size specified in the OP, but this article mentions sizes up to 900 x 450 x 20 mm (36 x 18 x 0.8 in). Sapphire has a hardness of 9. Artificial sapphire is clear (no color), and its index of refraction is only 1.77 so it doesn’t have the “sparkle” of diamond. So I suspect it’s not very attractive to scavengers.
Silicon carbide might be another good choice, but I don’t know if they are available in large solid pieces.
Over the time frame that we’d be looking at for these obelisks/steles, wouldn’t corrosion and oxidation be a much bigger concern than mechanical weathering?
I mean, if you put it somewhere prone to sandstorms or directly on the coast, then maybe not. But if you put it somewhere inland and not on a river or in a flood plain, there’s just not that much flying crap to really abrade it much.
Freeze/thaw cycles might be an issue if there are cracks or voids (no granite!), but I’d think over the long haul, solar exposure and general corrosion would be the issues.
Wouldn’t something like glass or maybe synthetic corundum be the way to go in that case?
That sounds a bit like the Georgia Guidestones, although they’ve already been defaced.
Think big, very big. Carve the information into the tops of cliffs on the Nazca Plateau or similar terrain. The larger the symbols and the deeper they’re cut, the longer it will last.
But for normal sized stuff nickel sounds like a good way to do it. Instead of making a solid block 3 inches thick stack up layers of etched sheet nickel coated in PTFE. The outer sheets could be thicker material deeply etched to be reconizable, the inner sheets can be a thin foil. You might want to include some sort of Rosetta Stone translation guide assuming any major current languages might survive. A similar guide for mathematical symbols should be provided. A simple diagram showing dots to match numbers and a series of simple equations to explain operators should do. Also, if you want people to pay attention to the material, include some porn. And just for laughs, a picture of a Jackelope.
Go big? Carve it in the moon! 
Fossilized human skulls. Carve your info on complete skulls and place them in tar pits, sedimentary ponds and such. Hell make it financially attractive that everyone that dies gets their skullen engraved to carry on information.What is the oldest biological stuff in museums, fossils. Future beings will end up collecting and studying them at some point. At first they will be a curiosity and saved until the civilization matures; then they will be sought after. Encode how to purify water and eliminate germs. That will get their attention.
Hell, encode how to fermate grains into alcohol. You will be worshipped as a god.
Why not just put the stuff in a place where it won’t face erosion? Maybe bury it 5 feet below the surface of grass somewhere.
Then you run into the problem of ‘how do you know where it is in 500 million years’. No idea. there could be a supervolcano, or some other giant issues that totally change the landscape.
Having said that, when reading about these issues I heard that our bronze statues and Mt Rushmore will last very long if humans disappeared.
So I would write them in bronze. I don’t know if Bronze is the hardest and/or most erosion resistant metal out there, I seriously doubt it. As others have said, there are programs using Nickel instead.
But carving them in metal seems like it would be pretty long lasting.
Those people suggesting fossils: the odds of fossilization are much lower than you think. In A Shory History Of Nearly Everything, Bill Bryson says there haven’t even been enough humans for a complete human skeleton yet.
How about aluminium magnesium boride as a material? - a hard, durable ceramic - very low coefficient of friction (which helps reduce wear/erosion). Not sure how resistant to chemical corrosion it is though.
The OP does not specify how long the data needs to last.
The Rosetta Project is supposed to be readable until about 10,000 years from now.
If our global civilization collapses in 2050 (and I wonder how that would happen, as do you all), how long would it take to reintegrate enough of a culture to decode something written? Would we preserve any written material? Would we revert to a totally verbal culture? Would we rediscover the ‘magic’ of written words soon if only because through our story-tellers, our descendants were told that the days of old held great wonders that were recorded on paper/stone/unobtainium?
What language would you use? Honestly, the material that the data is encoded with makes no difference if you have no touchstone, no connection to the words that you have been taught to use. Generations come and go, verbal languages change. What was written may be totally indecipherable in a mere few centuries.
But, I digress, obviously. The OP asked what we could use to make this monolith-o-knowledge. Remarkably, it appears that a nickel alloy works well, and is used by the Rosetta Project. If such an artifact were encased in a silicon dioxide sphere and then scattered to various places on the planet, that might well work. Make them a meter or so in diameter. Inscribe on the exterior some pictographs that indicated the interior had secrets from the ancients.
But then, again we come to the problem of written language. Pictographs. Hrrrrr.
I probably should stop thinking about this now.
Thanks for reading.
You recall wrong. Lots of rocks are harder than laterite, which is more of an oxide-clay soil horizon than a rock (and laterite is a mix of oxide minerals, what you have there is the formula for magnetite, specifically).
Laterite was used at Angor because it is, in fact, easy to cut into blocks.
If I were going to use a rock rather than an engineered material, I’d go with either diorite (see: Hammurabi, Code of) or orthoquartzite (metamorphosed sandstone composed practically entirely of quartz).
True, but that data is based upon random creation of fossils. What I am suggesting is that we take what we know about fossil creation and purposely and consistently apply the conditions to a large number of skulls. Fossil records of dinosaurs go back 240 million years, all other man made artifacts pale in comparison. If that fossil pictured had marking all over it we would be going ape shit trying to figure it out.
Also it would be best to use petroglyphs to communicate to the beings of the future; even though we may not be able to read these yet there are people working on it. Perhaps someday we will come across that first beer formula.
The problem with bronze is that people would figure out pretty quickly that you could melt down the useless tablets to make swords and armor suchlike. Bronze is to easy to destroy. You need something that is durable, yet useless except as a solid block.
So slabs of granite might be repurposed as building materials, but if they’re scavenging the slabs for building materials they aren’t likely to deliberately chip off the writing. Instead they’re more likely to mortar the slabs in place whole, and thus preserve them.
So maybe what we really want is to create millions and millions of largish granite or other durable stone blocks inscribed on all sides with useful information. Each brick itself won’t contain a whole library, the brick would just be a single page, but we’d create lots of them. Some bricks are language primers, maybe we could make these a special color.
We leave the bricks lying around in piles at likely future building sites, each pile with maybe a dozen or a hundred copies of each “page”. Future generations of people come across the bricks, look at the funny squiggles, then shrug and build houses out of them. Years later priests and gentlemen of leisure get interested in the funny designs that the peasants have on their bricks, and try to figure them out.
And the bricks have been preserved because they are useful exactly as they are, and don’t need to be transformed to make them useful. We don’t rely on single large monoliths that can be buried or lost, we rely on massive redundancy and ubiquity. And we don’t worry so much about the blocks being lost, because they are constantly being scavenged and reused, and even if lots of bricks get ruined or broken we still have plenty of backup.