Tonight's red post collection features another preseason developer blog on New Jungle changes coming in Preseason 2015, Riot Stashu with a set of follow up changes for the Cassiopeia rework, and Tryndamere offering insight on Riot's communication philosophy.Continue reading for more information!
- Red Post Collection: We are Vengeance, Vi's E Bug, Sion passive change in 4.19, and more
- Red Post Collection: Preseason 2015 Increasing Strategic Diversity blog, Updated Summoner Rift's new music, and more
Preseason 2015: A New JungleThe preseason information dump continues! While these changes will not be in the next live patch are still in need of additional PBE testing, Riot has put up a dev blog on the tentative preseason 2015 jungle changes!
"Fearless and Axes here to talk about the jungle changes coming in the upcoming 2015 Preseason. Our big goal this time around is to bring strategic diversity back to jungle. Currently a small pool of optimal junglers has been skewing team compositions toward a small set of strategic plans, so our focus for 2015 is to let players take the junglers they love and the junglers they want to fit into their team’s strategies. I realize jungle diversity improvements have been a focus a few times in the past (has it been every season?), but this time around we’re taking bigger steps and we’re excited to share our ideas with you.
Rebuilding the Jungle
As I mentioned in the foreword, the 2015 Preseason changes are taking a crack at some problems in the jungle that have built up over many seasons. In the past, our approach to the jungle has been through smaller sets of changes, aimed at fixing only the issues that arose in the previous year. Our fixes did largely address past problems, but they also didn’t have the mechanisms to deal with some of the power balances that appeared later on.
As players mature in their understanding of the game landscape, the strength of a jungler ends up getting boiled down and optimized into a small handful of champs who do roughly the same thing: who can clear the fastest while also having strong dueling / ganking capabilities? This season, for example, we saw a lot of Kha’Zix, Evelynn, Rengar, Elise and, everyone’s favorite, Lee Sin. A side effect with having ‘optimal’ junglers who crowd out other choices is that teams tend to orient their team selection and strategies around their limited choices. For example, If there aren’t any powerful junglers that work in a siege comp, it becomes that much more of a gamble to commit to one.
The 2015 changes are aimed at building a jungle that has components to empower and react to many different actions that junglers take. This allows us much finer control to react to the live state of the game with small changes to the jungle, rather than only having to tune champions and items. In past seasons, collateral damage with each jungle change was very high, where a buff to a few champions creates huge imbalances (anyone remember Feral Flare?). The new jungle aims to give us many angles to attack problems as they arise, to give us more precision than we’ve ever had before. More scalpels, less sledgehammers.
One of the symptoms of a dominant strategy is that players feel like they know the actions they’ll take throughout the game by the time the loading screen is up. After a few dozen repetitions of the Lee Sin vs. Kha’zix match up, we as junglers have a very good idea of how the match up will play out, and the game is very likely to feel very similar to the games before and after. Not only is this not delivering the level of excitement League of Legends should bring, but it takes the strategic decisions out of a role built around strategic choice.
Junglers are the role most free to make strategic decisions throughout the game. Not being bound to lane, they largely self determine their points of interaction on the map. The jungler role is most rewarding and healthy when the strategic decisions of the jungler are as important as their mechanical play. On the other hand, when there are far too few variables in each game, jungle play feels much more shallow and gets focused on tiny optimizations. Without the constant presence of a lane opponent, and without interesting decisions to make, the jungle position becomes less deep and diverse.
Given our high-level goal this season is strategic diversity, each of our following changes (reminder, they’re still in testing!) is subservient to that goal. Whether it be modifying the pace of the jungle, adding new jungle monsters, or giving buffs through Smite, all of the below changes are in pursuit of opening up strategic diversity in both the jungle and for teams at large. By injecting multiple sets of new variables, we can highlight the unique strengths of each jungler so that all have different optimizations and can interact with the jungle in more unique ways.
First, the current jungle is heavily biased in favor of fast-clearing junglers who double up on their sustain through offense. Our first goal was to balance this more equally in favor of other attributes like sustain and mitigation in order to open up the strategic space for jungler power. As such, we’ve given basic camps a pretty hefty base stat increase. We’ve also changed the pacing of the basic camps, increasing their gold and XP rewards, along with their spawn delay. This puts a higher premium on mitigation-focused junglers who can stay healthy to gank after clearing a camp, while high-damage junglers might be more effective at ganking but will have a tougher time keeping up.
We’d previously moved away from junglers being able to full clear their jungle as this meant they were then forced to take action on the map. With the increased power of counter jungling, the new crab camp (more on this in a minute) to take in the river, and the new vision tools from the camp rewards(almost there!), junglers should now have more meaningful strategic gameplay even when their camps are down.
This brings me to my next point: jungle camp rewards. Each jungle camp now grants a buff the first time Smite is used on them (this buff is available each time they spawn). We’ll have more details on these camps as we get them dialed in for release, but the goal of each reward is to provide a flexible tool that junglers can use in different ways based on the current state of each game.
For example, Gromp currently gives the jungler a poison armor that, you guessed it, poisons enemies that attack the jungler, dealing magic damage over time. This buff is great as a starting point to help tanky junglers in their first clear while also providing safety vs. invades as well as some power in team fights and objective control. Each jungler should have a reason to want each camp’s buff throughout the game, so that junglers are always making meaningful choices on what camps to take and how to tweak their routes to fit their larger goals.
Finally, let’s talk about the new Scuttle Crab camp. This is the new monster found in the river that actually runs away from its attackers. The goal here is to give junglers a new reason to venture out of their jungle and contest the crab in smaller skirmishes. Currently, when a team defeats the Scuttle Crab, they’re awarded a vision bubble and movement speed shrine that grants out of combat movement speed in front of Baron or Dragon pit. While the Scuttle Crab itself isn’t dangerous (but is durable), the chase for it can pull champions up and down the river. This means that taking the crab can easily turn into a small fight with a few champs on each side. There are larger team-based implications, like creating a new decision point for mid-laners to roam for vision and safety, but I’m focusing mostly on the jungle impact here.
We’re very excited to see how players use the Scuttle Crab monster, and what crazy strats shake out of this addition.
All of these changes are only half the equation. Let’s talk new jungle items! Axes, I choose you!
Jungle Items: Now With 40% More Smites!
Preseason will include a completely new set of jungle items, built from scratch towards a different set of goals from previous seasons. This is obviously a big change, and before getting into the details, I want to take a minute and talk about why a large-scale item change was necessary in the first place.
The previous jungle items combined your champion’s role (Fighter, Tank, Assassin) with a fairly specific pattern of jungling. Spirit of the Elder Lizard, for example, is an item that’s only good on AD champions dealing sustained physical damage - mainly Fighters. It’s also only good on champions who want to gank or duel before moving back into their jungle to heal up and claim their bonus gold.
While these results correspond with our goals from last season, we saw two problems over time: first, champions whose have a playstyle that doesn’t work well with their optimal role item are left out in the cold. Shyvana is never going to be a great Spirit of the Elder Lizard user because Shyvana doesn’t need a mechanic that stores up gold while she goes and does something else. Instead, Shyvana stays in the jungle pretty continuously because she has a kit that empowers her to safely and quickly clear her jungle and invade but leaves her as a relatively lackluster ganker.
Second, our previous items amplified role problems. When fighter junglers get an item that makes them particularly good at fighting enemy champions as their core; jungle purchase - and tank junglers don’t - then whenever that item is strong, fighters will tend to stomp tank junglers out. When that item isn’tstrong, fighter junglers just don’t get picked at all because they have no real fallback item. Fixing this problem doesn’t instantly level the playing field among all junglers - it doesn’t make Sejuani a top tier jungle pick by itself - but it does mean that our jungle items are no longer exacerbating her problems.
There’s also an opportunity here: the new jungle is about diversity of action. In addition to the ubiquitous ganking of Season 4, invading the enemy jungle has become more valuable, there are new and more impactful objectives to contest, and there are more ways to engage in the vision war on the map. Even if you pick the same champion in two consecutive games, the conditions of those two games should result in different experiences. If we can create a set of jungle items which support different important jungler actions (depending on circumstances) then these items can serve as a good tool to help promote this diversity of action in the jungle.
On to the items! We’re retaining Hunter’s Machete as a starting item, but changing how it deals damage and sustains your champion to be nearly role-agnostic - in other words, it shouldn’t be better on a fighter than on a tank. Hunter’s Machete then upgrades to four other items, each of which changes your Smite spell so as to augment what type of action you might want to take as a jungler.
Stalker’s Blade increases your power while ganking, allowing the jungler to play a larger-than-usual role in helping snowball lanes and put the whole team ahead.
Poacher’s Knife provides you with additional power and incentive to cross the river and invade the enemy jungle, stealing whatever camps you can opportunistically take and putting the enemy jungler behind.
Skirmisher’s Sabre gives combat power specially targeted to a 1 versus 1 situation, empowering you to seek out the enemy jungler and try to kill them, to show up during enemy ganks and turn the tide of a battle, or to punish an invading enemy jungler.
Ranger’s Trailblazer makes you faster and safer at fighting jungle monsters, allowing you to stay safer while farming because your health will be generally higher and giving you more time to take other actions.
These items are role agnostic - they can be purchased whether you’re a Fighter, an Assassin, a Tank, or something else (jungle Karthus?), as they don’t provide stats binding them to one role or another. The only requirement is that the action they help you to do has to be important to you as a jungler, whether as a strength that you want to reinforce or a weakness that you’re looking to shore up.
These are items, however, and to want them in your full build they probably need a good set of stats! We’ve also addressed this issue by creating an enchantment system for them to provide stats relevant to your roles. The Warrior enchantment adds attack damage and Lifesteal, the Magus enchantment gives ability power and cooldown reduction, the Juggernaut enchantment adds health, armor, and tenacity, and, finally, the fourth enchantment, Devourer, gives attack speed and on-hit magic damage which stacks up as you kill enemy champions or large monsters. Each enchantment gives roughly the stats you’d get for the total cost of the enchanted item, so the jungle passive from Hunter’s Machete and the additional bonuses on the second tier items come effectively for free on the final items.
That’s it for now! Big changes are coming, and we're excited to take this next step into the jungle's evolution with you."Looking for more on the upcoming Preseason 2015 changes?
As a follow up to Cassiopeia's gameplay update in patch 4.16 and all the player feedback he's received since then, Riot Stashu has posted up a tentative changelist for Cassiopeia that he'll be testing on the PBE soon!
"Hey guys! Today I’d like to go into some detail about our upcoming Cassiopeia changes. Real quick though, I’d like to add that I’ve been surprised to see all of the attention Cassiopeia has been getting on Boards-- I’m really happy to see that so many players care about our favorite Snake Lady, even though things got a bit... passionate. Your enthusiasm is definitely noted.
Also, sorry if I haven't been as involved as I should have been in directing the conversation. At the end of the day, I had to take a step back to examine all of your feedback as I distill it into something actionable. Truthfully, it felt like many of the discussions would spiral into the same 10 paper kits or unconstructive comments on repeat, which wasn’t particularly useful. I’d like to get into deeper debates about why a proposed spell will or will not work, but those can be pretty tough because they usually don't just stop at just one reply. That said, I’ve been reading virtually everything, and there has been a wealth of insightful and helpful feedback aside from these, even since the beginning.
So, let’s get into it. Here are a few of the major pain points that many people seem to share:
- Cassiopeia is no longer a poison mage
- While this sentiment is a bit harsh, I totally agree that we’ve lost some of what made Cassiopeia so cool to many of her biggest fans. This is my biggest regret from the update, and we’re looking to get this back.
- Aspect of the Serpent is chore-ish
- This was not the intent. The most chore-like part (minion farming) has too much focus for what was intended as a fall-back, so we’ll be addressing that as well.
- I miss the long range, AoE, DoT pattern
- In actual fact, DoT’s do a lot more damage than anyone realizes, and often times so do AoE’s, so in order for Cassiopeia to be balanced as a long-range AoE DoT dealer, she’d have to take some significant hits. We had to make a choice here, and we went with giving her the most potent DoTs on single targets. Cass can still apply DoT from long range, but accessing her full power requires commitment to a single target (with E). This way we can give Cass meaningful DoT power on single targets instead of weak DoTs to all enemies. Fishing from long range with Qs and Ws is an important and fun part of her play pattern that we want to keep.
So with all that in mind, let’s take a look at a proposed changelist (which should be hitting PBE soon):
- Aspect of the Serpent
- No longer stacks on poisoned unit kill or on Twin Fang casts on enemy champions.
- Now, stacks once every 6 seconds (10 times per minute) and for every second that an enemy champion is poisoned.
- Breakpoints adjusted to keep them at approximately the same game time, but this is still up in the air (i.e. maybe they should come sooner).
- Noxious Blast
- As it turns out, this ability uses old tech and has always been inconsistent in delay, hitting anywhere between ~0.25s and ~0.5s after casting. This means that the ‘change’ to a 0.3s delay didn’t actually do anything >.> We’ll be fixing this is the same way that we’ve fixed abilities like Rupture and Dark Matter in the past, so now it will consistently hit at the same time. We’re starting with a 0.35s delay time, which should ultimately result in a much more reliable ability (this is a buff), but will see if we need to change this.
- No change.
- Twin Fang
- No longer restores mana on unit kill (mana costs reduced accordingly). Some people really enjoyed this, so I’d consider adding it back, but with the minion farming mini game gone, this seemed less fitting on the kit.
- Now adds a debuff on the target which significantly amplifies further poison damage. We have less confidence in this change, but it’s potentially super cool and solves some problems, so I can’t wait to see how it tests. Also, this is strictly a buff.
- As a note, this is not possible yet, but we're looking into some better buffering tech on spellcasts so that spamming the E button as it's coming off CD is more responsive.
- Also, how does everyone feel about the current CD? 0.5 s might just be a little low, as it keeps Cassiopeia locked in place, casting E's and hardly being able to move. I'd love to hear what people would think about skimming off some power here (say, bringing it to a 0.75s CD or something) and loading it into the DoT amp aspect, or something else. Thanks!
- Petrifying Gaze
- No change.
And there you have it! Please let me know if there are non-power related kit concerns that you feel are unaddressed, new problems created, or things you miss about old Cassio that these changes don’t bring back.
By the way, in these discussions, it can be really hard to get through proposed solutions that are focused on number analysis or are of the form "here's what the ability should do!" They get focused on what's difficult to solve or what might create more problems, rather than being focused on honing in on the original problem -- the real purpose of the discussions in the first place. Personally, I find the most valuable responses to vaguely fit some form like “I don’t feel that [change] solves [problem] for[reasoning],” or “These changes don’t address [problem], which I believe is concerning because [reasoning].” This lets both of us discuss possible solutions and hash out potential problems rather than arguing about the validity of suggestions :)
Lastly, as, a disclaimer, none of the above is certain, and everything is still in testing. But you can expect to see most of this or some form of it hit PBE pretty soon (probably early next week)!
Alright, that’s about it. I look forward to hearing what everybody has to say :D"
When asked if the new E buff to poison damage also included the damage from E itself, Stashu noted:
"Nope, E is not considered poison damage. Currently, neither are other sources of poison (Teemo, Singed), the way it is hooked up. I can change this part, definitely considering it, but E will not be poison damage."
Tryndamere on Communication Philosophy (& AMA's)In response to comments in a a recent AMA request thread on reddit, Marc "Tryndamere" Merrill, the President and co-founder of Riot Games, jumped on reddit to share a look into Riot's communication philosophy:
"I just read the AMA request thread for me and Ryze and thought I'd take a moment to kick off a new thread to summarize some caveats and orient people towards our previous experience with AMA’s and how we think about them.
A bunch of comments in that thread were accusing our previous AMA's as being "coached" and/or "PR bullshit" and I thought I'd take a moment to provide some background context on how we (I) view communication in general with our player base and with Riot as a company - because I’ve always been confused by those comments.
1) We encourage all Rioters to be themselves and to interact directly on the forums, social media, etc. We like interacting as part of the community and don’t see ourselves as “above” it, but we recognize the important role that Riot of course plays in helping to shape and influence it.
2) As a result, as individuals who love League and games in general, we tend to prefer “informal” communication and interaction and the more “formal” or “official” it becomes re: Riot / League, the more polish and thought that needs to be put into any communication. This is why we and many Rioters will drop into threads and contribute as parts of the community, whereas when we launch new content, features or make announcements about anything (like eSports), there are a lot more considerations that need to be taken into account to ensure that those messages (which typically have a lot more reach) don’t lead to incorrect takeaways.
3) We are all now accustomed to the fact that we have a very diverse player base (with tens of millions of players) and as Abe Lincoln said, "You can please all of the people some of the time and you can please some of the people all of the time, but you can't please all of the people all of the time" - no matter what we do, we let some people down. Even if we please 99% of our playerbase, that 1% is still hundreds of thousands of people. And as people who care about trying to do a great job to satisfy and entertain you all, when we peceive we have let anyone down (even if it’s a small minority), it still sucks a lot to see the problems we cause or missed opportunities to deliver awesomeness.
4) Because we know we can’t always please everyone, we have even more conviction to encourage all Rioters to “be themselves” and for us to “be ourselves”. Why bother posing? If someone doesn’t like our stuff, well, sorry whatever we did didn’t resonate with you, but what we do does tend to resonate well with an awful lot of core gamers, and seeing people delighted by the experiences we create is quite motivating and is the thing that gets us all out of bed every day. Per the above, we can’t please everyone and we don’t want to even try to (my mom & dad will never be League of Legends players - and that’s OK!)
5) I assume that a lot of the frustration that happens from some of our previous AMA’s is because of the thousands of people who have questions, we can't ever get to even a meaningful percentage of the different topics / questions people have, and thus many feel that we dodged, or they're just not satisfied with the answers. IE - east coast servers where?!!?1 When we’ve already talked about something, we likely just rehash the latest information that we’ve provided, because that’s the most relevant info. So if anyone asked about US latency and connectivity now, we’d probably say something very similar to this: (http://boards.na.leagueoflegends.com/....)Additionally, we simply won’t announce new features or spoil cool upcoming content, games or events in an AMA. Why? Because it actually would be a let down for you guys to get answers to some of those questions rather than letting us take the time to setup a cool launch campaign and/or do the “thing” that we have been working on for a long time justice. If we talked a ton about Jinx and the music video before launching her with the graffiti on the website vs. Vi and then launching the video, then launching her spotlight, login screen and turning her on in game, well then the excitement level would have been way lower and she would be less iconic and cool. Why would we rob anyone of that potentially cool experience?
6) If people ever want more random & off the cuff stuff, you can always follow us (or other Rioters) on twitter https://twitter.com/MarcMerrill &https://twitter.com/BrandonBeck or read any of my handful of blog entries: http://tryndamere.blogspot.com/ - it's not like we're hiding or we have any PR group edit the random stuff we post (like this thread, no one “reviews” this stuff”)
7) Anyway, back on the topic of our previous AMA's - we really do try to talk honestly about every topic because we have nothing to hide. Will we leak upcoming things though that we are working on? No (see #5). So then the question becomes, what can make an AMA interesting (not just with a Rioter, but with any AMA)? Well, that's up for you guys to decide, but here are some suggestions:
Remember that these are human beings that you are talking to on the other end of the interwebz. They lead full lives, have families, political views, are subject to bias, enjoy certain hobbies, have aspirations, have challenges, and have often had really interesting life experiences. Maybe you could get to know the person (if interested) and try to benefit from their perspective? If you're talking to someone like Tom Cadwell (Zileas), well, he happens to be a pretty smart guy, maybe he can give you guys advice on how to think about your education or career in addition to sharing perspective on design philosophy. If you're talking to an artist, maybe you can ask what inspires them or got them into art or what they love most about their work? If you're talking to Pendragon, maybe you could ask him about his work at Zappos (another very customer-centric company) previously or about his years of work helping to build the foundation for the Dota community (/duck).
Ask about “why” or “how” not “what”. It’s way easier for any of us to explain the rationale behind why previous things have been done or what our philosophies are that will guide our actions in the future than it is for us to talk about “what” will happen in the future. Additionally, the reason we make a bunch of “behind the scenes” videos and did the documentary series “Road to Worlds” is because we’ve very interested in providing you guys a window into how the sausage gets made. Oftentimes, it’s pretty interesting stuff.
8) Additionally, Riot can and will make mistakes. So will each of us as individuals. Do those mistakes define the institution that is the company? We hope not. We want Riot to stand for a lot more than the actions of one individual or one mistake - so we will correct our mistakes or own them when they happen, even if it takes us a long time to fix them. That’s one of the reasons Rioters would have to be exited from the company, or players would have to be exited from the community. Those things are larger than any one individual. One recent example that I’m excited about is the design a champion RAF issue (http://www.reddit.com/r/leagueoflegends/comments/2jc7b2...) where we’re trying to “tighten up” errors we made in the past with over promising and under-delivering.
Anyway, we’re all constantly working on making our skin ever thicker, but the criticism that is motivating to us that helps us improve the most is that which is rational, well-informed, constructive and gives us the benefit of the doubt from a motivation perspective (not assuming we’re actively trying to screw someone). And interestingly enough, that advice won’t just help you get through to Riot to help us improve more, but would likely serve you guys well in any relationship you have in life.
Anyway, off to take Linken to watch some Trick or Treaters. Have a happy Halloween and hope you guys are enjoying the Harrowing.
TLDR: We try, communication is hard, trust our intentions, even when we make mistakes & relax, Rioters are real humans too ;)"
As for what Tencent's input is on Riot games, Tryndamere explained:
"No hand in day to day. They just publish LOL in China (it's illegal according to Chinese law for a non-Chinese company to publish a game in China) and are investors (sit our Riot's board like our VC's did)."
In response to criticism of a "Do-something-wrong-and-apologize" route, Tryndamere commented:
To be fair, with all other companies in the world and the way they handle public critisim, it's not weird to accuse other companies, including riot, of this 'pr bullshit' you are talking about. In a way, a lot of the answers given are easy to generate even as a non-riot employee.
Anyway my problem with riot is that they take the 'Do-something-wrong-and-apologize'-route. A lot of things could've been handled in a better way, but in the end we get this public apology for you guys not having communicated well etc.We'd like to not make the mistake the first time around too. But when you do make a mistake, what are your options?
1) Own it and apologize and try to improve
2) Deny it, obfuscate or don't respond
We'd rather do #1 if we ever do make a mistake. And remember, growing a company from 60 people to over 1500 in 4 years to serve 15 different languages and tens of millions of people is not easy.
But, I think we can all agree we keep getting better, and that's what is most important IMO. Failure is a key part of learning - people and companies who are unwilling to acknowledge their failures and unable to learn from those mistakes (so teach your children that it's way better to embrace mistakes and work through the challenges to learn how to overcome them)."He continued:
I thought companies of riot's calibre would invest enough in whatever is needed to not make any mistakes at all. I was under the impression that throwing money at the problems would solve most things and I think some others feel this way too.Money doesn't solve problems - people do. And it takes time to groom, train and build teams of people capable of solving complex problems or to come up with great creative solutions / ideas."
Tryndamere also answered a question regarding communication education for new Rioters, explaining:
"Yah, we have "red name" training that Rioters need to go through to earn their Riot badge to post as a Rioter. It's a short class where we explain things like:
Remember, almost all Rioters have previous experience at other companies where things are very different so we have to help them learn new, positive habits."
- Be yourself
- Don't leak / hint / hype stuff we're working on (& don't hype announcements)
- Focus on your area of subject matter expertise (IE - Dr. Lyte focusing on player behavior, etc) because you know it best and it's better to not speak for others
- Be respectful to our players
As for the intent behind some of the more aggressive or even insulting feedback, he shared:
We are verbal, and might seem offensive, cause we truly love this game. So just wanted to say thanks for everything."Yah, we know ;) I used to sit in your guys shoes and pour my heart out on SOE's message boards complaining about Everquest with a litany of suggestions on how to make the game better.
The reason I did that was because I loved the game and I wanted to help.
We know you guys feel the same way and it's appreciated ;) I'm just trying to take a moment to remind you guys about how you can best help us."
As for his thoughts on how Riot handles toxic players, Tryndamere noted:
What is your opinion on reformation and the current status (or lack of status) of the way toxic players are dealt with in the average players game?
We've made a huge dent that will continue to get better. Brand trails execution though, so we have to continue to get better and over time, the perception re: negativity will continue to change for the better."
Tryndamere also answered a question regarding upcoming security improvements:
"We're working on a lot of security features - this stuff is important and takes time to get right.
We had to refactor a lot of how our whole account system and platform worked to make it more secure and will continue to be a big priority."
When asked why he didn't post the same thing over on the official Riot boards instead of only putting it up on reddit, Tryndamere replied:
"I posted there because it was in response to a thread that was located there and I wanted Redditors to see it.
Your thoughts are correct - with Boards, we'll be able to do lots of cool stuff here."He continued:
"Oh come now - if we didn't care about the Boards etc, we wouldn't have spent a year + creating them ;)
We're just getting started with them."