1. Self-administering gene therapy will become a more prominent, though still underground, thing. I know a few people who have already. Exciting! 2. Self-admin

1. Self-administering gene therapy will become a more prominent, though still underground, thing. I know a few people who have already. Exciting! 2. Self-administered gene therapy will quickly be made illegal in at least one U.S. state. 3. Mainstream technology will continue being very boring and decades behind research and what hobbyists are doing unless intervention happens (think CARDIAC on a wider scale and a more broad dive through the dumpster of technology). 4. We're going to see a Linux operating system that's actually pleasant to use. Not just a window manager, not just a desktop environment, but an operating system. It probably won't go mainstream, but it's going to be there. It may use GNU tools, it might not. It'll probably have its own suite of programs. 5. Systems software research is going to return, even if it's not going to be as widely honored as Plan 9. I think one thing we're going to see is at least one person pursuing making a complete system in a minimal amount of code: think FoNC, but probably smaller. 6. IRC will still be in daily use by at least a few thousand people. 7. Rust won't take over the world. 8. Bryan Cantrill is going to be a lot richer yet still virtually unknown. 9. illumos is going to see a minor resurgence, a bit like what Plan 9 has experienced, yet a bit more commercial. 10. YCombinator will probably get better now that the leadership has changed in a positive direction. (Reasoning: very soon after the most recent change in leadership, they stopped bending to China. This makes a person hope that the trend to not maximize profits at all costs continues.) 11. The surveillance state will intensify. 12. It won't intensify in a way that causes meaningful action on behalf of legislators. (I'd love to be proven wrong in this.) 13. Paul Graham will probably come out of his self-induced exile/pseudo-retirement. My bet is that it will come before his kids are fully grown, though thoroughly after his essays finish the decline in quality we've seen since 2012 or so. (Not sure if essays will return to former quality when this happens.) 14. The political climate will be much different. I'm not sure whether this will lead to meaningful change in how elections are executed. 15. VR and AR will not meaningfully catch up to the MR research of 2005 for some time. 16. One of the acquisitions that happens in the 2020s is going to hurt. "Oracle buys Sun" level of hurt, but worse. 17. Adolescents will continue to be infantilized. 18. IBM will still maintain compatibility with OS/360. 19. IoT devices will be breached en masse after it hits a critical point. 20. The amount of vulnerable routers in America is going to hit a breaking point. (With a bit of luck, this will be used for good, not evil. Possibly distributed Tor relays?) 21. Political tracking is going to become far more prominent, and far creepier. With how much data you can get by going into virtually any government office in America with $50, it seems inevitable. Might be under the radar, though. 22. HN will not look meaningfully different when accessed via news.ycombinator.com on a conventional web browser. Maybe if we're lucky we'll see higher resolution arrows (or SVGs...pleaaaaaaaaaase, 'dang?), but I don't think we can count on that. 23. The increase in cost of consumer goods because of the recent punishments on Chinese imports will make the lives of poor American consumers tangibly worse. 24. APL might not see a comeback, but it's going to be a lot easier to learn soon. 25. Taxes on sugar-filled goods will be demonstrated to have an actual, tangible impact on the average weight. 26. We will not see as large of an increased (percentage-wise, not byte-wise) in storage capabilities for your average consumer until at least 2028, probably not until after the decade is over, though. 27. Xerox will buy, merge with, or be bought by H.P. 28. Tech journalism will continue to get worse. 29. There still won't be a language that's as easy to learn as FORTRAN was in the 1960s. 30. Copyright terms will be extended again. 31. We're going to see the first deaths of notable HN users that aren't by suicide. I'm pretty sure there are multiple people on /leaders that are older than 60. 32. 1999's ThinkPads will still be humming just fine by 2029, though they'll probably need their batteries replaced. 33. We're going to see the return of real hardware companies. 34. A Linux distribution will ship with nice default fonts rather than the poor ones that are shipped commonly now. 35. TAoCP will never see itself finished, sadly. 36. Linus will continue getting increasingly conservative in what he allows into the source tree. (He'll probably step back before it gets too bad, but I don't feel confident to make a prediction on whether or not that will be in the 2020s.) 37. Struck with the realization that 2020+N seconds is closer to 2050 than 1990, we're going to see at least a few notable tech founders very visibly and noisily adopt religion. 38. 'idlewords will continue his metamorphosis into a gonzo journalist. 39. It will be magnificent. 40. OpenBSD will see an increase in adoption because of its history of sane choices and solid reputation for noticing security catastrophes before they happen. 41. C will still be prevalent. 42. Unfortunately, Python will still be top dog in the ML space. 43. Interpreted languages will experience a renaissance. 44. 1366x768 is still going to be in the top-10 resolutions globally barring the event that VR displays become prominent. 45. Algorithmically-generated music is going to become cheap to generate, Spotify and YouTube will get in on it, and weight recommendations for all but the most niche listeners to whatever their algorithms are putting out. 46. 9front will still actively get improvements. 47. Electronic mail will still be used frequently, and the ecosystem will decay even further. 48. Traditional companies will become tech companies to an even greater extent as they notice their revenue peaking, and if there's any chance of a research operating system becoming a global phenomenon again, it will be because of this. (I think that revenue has a chance of aggressively hitting and falling from its peak because of falling birth rates in countries with currency that's worth anything.) 49. In the event of venture capitalists becoming more conservative, stack will begin to matter again as the absurdly high cost of the cloud becomes more meaningful. (Whether this manifests as a resurgence in efficient languages or a resurgence in dedicated hardware I'm not sure of.) 50. Unless Maciej gets hit by a bus (back up your bookmarks!), Pinboard will continue its brutal and unending conquest of tech.