Tree Line is one of the best barriers printed recently. If you're looking for a barrier to keep the Runner out – as opposed to simply being a gear check, or something used to gain economic advantage and pull off combos, or something intended primarily to delay the Runner a click and waste one run's worth of credits – then Tree Line is probably your best choice in the current metagame.
First off, this ICE's basic statistics are very good. Normally, mid-sized barriers are strength 3, cost 3, and have some minor upside: Maskirovka, Klevetnik, Masvingo. (The original of course was Wall of Static, but ICE is slightly better than that nowadays. Palisade is also an interesting comparison.) Tree Line has strength 4, which makes it 1 more taxing than your typical mid-sized barrier; it also costs 1 more to rez, but you will get that credit back if you can fire the subroutine (and you frequently will be able to fire it on the initial rez). For Cleaver, one of the more commonly used fracters now that Paperclip has rotated, it costs 2 (or one Leech counter) more to break than your typical mid-sized barrier. These are not bad stats at all, and would mean that glacier decks would be seriously considering this ICE anyway (even out of faction) – given that Fire Wall is no longer legal, your next step up in strength would be Pharos which is considerably more expensive.
With the release of The Automata Initiative, and the rotation out of the Flashpoint cycle, barrier strength has suddenly become much more relevant than it has been in the previous two metagames. When faced with ICE that's a point stronger, a Paperclip would just pay an extra credit, and a Botulus or Boomerang (or the recently banned Endurance) wouldn't even notice. But Paperclip has rotated out, and there's now much more variety in decks with respect to how barrier-breaking is handled: Arissana frequently tries to break barriers with Slap Vandal (and maybe Poison Vial), several runners are trying out Banner, and the "default" fracter now seems to be Cleaver, with Propeller seen occasionally and Curupira a little more often. These newer, or newly-relevant, ways to get through barriers tend to struggle when they have high strength; Cleaver is hard to pump, Curupira has to be pumped quite a distance, and most of the rest have a limit for the strongest barrier they can break, normally around 5 or 6. This means that ICE that can be advanced for strength is in a fairly valuable spot in the metagame: if you can get the ICE above a certain strength breakpoint, the Runner's rig might have no way to deal with it. (Most commonly, the Runner is left with a single out, usually either Boomerang or Botulus; these cards are both good against Tree Line but it is often possible to overload them.) Tree Line is pretty useful in this respect because it isn't that far off the magic 6–7 strength that suddenly stops half the opponent's breakers working; you could get it up there with 2–3 and 2–3, which is expensive, but not completely unaffordable if it makes a server impossible to run.
I've been trying out Tree Line in pretty much all my Corp decks recently, initially in "let's try this out and see how it goes, I might want to write a review" mode rather than because I necessarily thought it deserved the slot. Generally speaking, it's been very good at what it does – frequently it's the card I'm hoping to draw, and when I do install it it does its job better than the alternative barriers would. I've found myself adding more and more copies of it over time, often starting as a 1-of and going up to the full 3. However, despite this, it isn't actually helping to win that many games. Part of the problem is that you can only run 3 Tree Lines in a deck, and the card seems to be hit by the 3/deck limit much harder than most cards are. Frequently it is hard to draw them by the point at which you need them. When you do draw them late, they're more vulnerable, because this is the general nature of ICE that's drawn late: if you can't put them on the inside of the server then they become more vulnerable to Inside Job, and Hippo, and Tsakhia "Bankhar" Gantulga. That's a downside that hurts almost all ICE in theory, but when you're running mediocre interchangeable ICE you don't notice much – it just starts really stinging with Tree Line because you know that some of its potential is being wasted. Part of the problem is that locking the Runner out of one server does not win the game; you generally need to defend at least two servers (typically R&D and a remote server; for some decks, this is instead R&D and HQ), and getting two active Tree Lines takes longer than getting one.
Tree Line has a second mode: you can pay , 1 and expend it in order to triple-advance a piece of ICE. This ability is in theory very good if you want to triple-advance a piece of ICE (it is slightly better for that purpose than Dedication Ceremony, because it works even on unrezzed ICE). You are spending one card, but saving two credits and three clicks; this is a huge economic swing if triple-advancing a piece of ICE is something that you actually wanted to do. There's a basic problem with this mode, though: if you're in a gamestate where you want to have highly advanced ICE, it is probably also a gamestate in which using Tree Line as a piece of ICE would be particularly good. So expending a Tree Line is a bit of a weird thing to do; if a highly advanced ICE is valuable enough against the current opponent that you're willing to spend resources on setting it up, it usually makes even more sense to pay the extra clicks and credits to install the Tree Line protecting a weak server and advance that manually. This means in practice that expending a Tree Line is a bit of a desperation measure – something you do when you need to reinforce your servers in an emergency, at the cost of having good servers later in the game – and so far I've lost every game where I've had to do it.
The expend on Tree Line is therefore probably best (only?) used in situations where having an advanced ICE is important, but having an advanced Tree Line in particular wouldn't help. There are a few ICE that work well advanced in certain specialised scenarios: Colossus (versus fixed-strength killers like Mimic and Num), Hortum (versus AI breakers like Aumakua and Audrey v2), and Oduduwa (in front of advanceable ICE to protect a server, usually R&D but occasionally HQ, from a deck designed to spam runs on it). Incidentally, Tree Line is a pretty good choice of advanceable ice to put behind an Oduduwa in order to make use of the free advancements; doing this is punished by Botulus but pretty much nothing else, so you only have one card to play around. These situations do come up, but they're rare, and in practice you never seem to have the Tree Line in hand when you need it.
So, in conclusion: if you're trying to build a deck around keeping the runners out of your servers, and are looking for a barrier to slot into it, Tree Line is going to be better than its competitors – but you are going to have trouble drawing it when you actually need it, and the Runner may have won by the time you can set it up. If you're looking for a mid-strength barrier to tax the runner a little, Tree Line may still be the best in slot now that IP Block has rotated. It's a good piece of ICE – although it improves the type of deck that may struggle for other reasons, and may not improve it by enough for it to be worthwhile playing a deck like that. It has an expend ability, but it's hardly useful because you have to choose between an expend and an install, and the install is almost always better than the expend: both are situational, but for most situations in which the expend is good, the install is even better. But even just looking at it purely as a piece of ICE, it's pretty good – against most decks, you don't even need to advance it for it to be good, its base stats are perfectly adequate on their own.